Proceedings of the First International Symposium on Trichoptera Proceedings of the First International Symposium on Trichoptera

Lunz am See (Austria), September 16-20, 1974

Edited by HANS MALICKY

Dr. W. Junk B. V. - Publishers - The Hague 1976

Conference photograph: 1. P. D. HILEY, 2. M. I. CRICHTON, 3. Mrs. MARLIER, 4. O. S. FLINT, JR., 5. S. D. SMITH, 6. G. MARLIER, 7. C. DENIS, 8. J. LE LANNIC, 9. M. BOURNAUD, 10. D. LHONORE, 11. Mr. LHONORE, 12. L. S. W. DATERRA, 13. Y. BOUVET, 14. M. MARINKOVIC-GOSPODNETIC, 15.1. FLORIN, 16. A. NIELSEN, 17. A. NEBOISS, 18. H. H. Ross, 19. M. FEY, 20. J. ILLIES, 21. N. V. JONES, 22. F. CIANFICCONI, 23. S. UJHELYI, 24. G. P. MORETTI, 25. M. H. HANSELL, 26. W. WICHARD, 27. R. M. BADCOCK, 28. H. ZINTL, 29. M. I. VIGANO-TATICCHI, 30. J. O. SOLEM, 31. H. MALICKY, 32. A. ADLMANNSEDER, 33. B. STATZNER, 34. F. VAILLANT, 35. J. C. MORSE, 36. G. B. WIGGINS. Not in photograph: L. BOTOSANEANU, B. HIGLER, V. H. RESH. Not present, but submitted papers: K. SMART, J. B. WALLACE. Photograph by J. T. MALICKY. ISBN-13: 978-90-6193-547-6 e-ISBN-13: 978-94-010-1579-0 DOl: 10.1007/978-94-010-1579-0 © Dr. W. Junk b.v. - Publishers - The Hague 1976

Cover design; Max Velthuijs Filmset in Northern Ireland at The Universities Press (Belfast) Ltd. Contents

Preface IX List of participants XI

Systematies and evolution (Chairmen, NIELSEN, VAILLANT and Ross) HERBERT H. Ross: Observations on the Helicopsychidae of New Caledonia p. 1 STAMFORD D. SMITH: A progress report on the phylogeny of Rhyacophila larvae p. 5 GLENN B. WIGGINS: Contributions to the systematics of the caddis-fly family Limnephilidae. III: The genus Goereilla. p. 7 PETER D. HILEY: The identification of British Limnephilid and Sericostomatid (s.l.) larvae p. 21 F. VAILLANT: Some Philopotamidae from France p. 25 JOHN C. MORSE & IAN D. WALLACE: Athripsodes BILLBERG and Ceraclea STEPHENS, distinct genera of long-horned caddisflies p. 33 LAZARE BOTOSANEANU: Communication sur trois larves de Trichopteres du Nepal (Progress report) p. 41

Zoogeography (Chairmen, NIELSEN, VAILLANT and Ross) ARTURS NEBOISS: A progress report on the endemic element of Tasmanian Trichoptera p. 45 OLIVER S. FLINT: A preliminary report of studies on neotropical Trichoptera p. 47 RUTH M. BADCOCK: The distribution of the Hydropsychidae in Great Britain p. 49 LAZARE BOTOSANEANU: Les Trichopteres de l'espace carpato-balkanique, fournisseurs de documents pour l'etude de l'evolution p. 59 HANS MALICKY: A progress report on studies on Trichoptera of the Eastern Mediterranean Islands p. 71 MARA MARINKOVIC-GOSPODNETIC: The differentiation of Drusus species of the group bosnicus p. 77 G. P. MORETTI, A. VIGANO & M. I. VIGANO-TATICCHI: Some informations on the orobiontic fauna of Trichoptera of the Italian Western Alps above 2000 m p. 87 G. P. MORETTI & FERNANDA CIANFlCCONI: The taxonomical and chorological problem of Drusus improvisus McL. in the North-Central Italian Apennines p. 93

Ecology (Chairmen, CRICHTON, MORETTI and Ross) YVETTE BOUVET: Ecologie et reproduction chez les Trichopteres cavernicoles du groupe de Stenophylax p. 105 G. P. MORETTI, F. CIANFICCONI & Q. PIRISINU: The Trichoptera population of a temporary ecosystem of the Umbrian Apennines (Perugia, Italy) p. 111 NEVILLE V. JONES: The Trichoptera of the stony shore of a lake, with particular reference to Tinodes waeneri p. 117 NEVILLE V. JONES: Studies on the eggs, larvae and pupae of Tinodes waeneri p. 131 JAMES BRUCE WALLACE: A progress report on the North American Macronema larvae: their retreats, food and feeding nets p. 145 MICHAEL I. CRICHTON: The interpretation of light trap catches of Trichoptera from the Rothamsted Insect Survey p. 147 ANKER NIELSEN: Pollution and caddis-fly fauna p. 159 ANKER NIELSEN: Revision of some opinions expressed in my 1942 paper p. 163 VINCENT H. RESH: Changes in the caddis-fly fauna of Lake Erie, Ohio, and of the Rock River, Illinois, over a fifty year period of environmental deterioration p. 167 vii Morphology (Chairman, VAILLANT) WILFRIED WICHARD: Morphologische Komponenten bei der Osmoregulation von Trichopterenlarven p. 171 BERNHARD STATZNER: A progress report on studies on the functional morphology of the genitalia in three new species of Cheumatopsyche p. 179 Behaviour (Chairman, MORETTI) MICHAEL H. HANSELL: A progress report on some approaches to the study of larval house building with particular reference to Lepidostoma hirtum p. 181 KEENAN SMART: A progress report on the building motivation in the caddis larva, Lepidostoma hirtum p. 185 HERIBERT ZINTL: House building: Problems about the spontaneous change of the architectural style in the larva of Potamophylax latipennis p. 187 MICHEL BOURNAUD: A progress report on the locomotion behaviour of a larva of Lim- nephilidae (Microptema testacea) in water currents p. 203 JOHN O. SOLEM: A progress report on diel rhythmicity in Trichoptera p. 205 Author index p. 207 Subject index p. 209

Vlll Preface

In past years there have been several unsuccessful attempts to arrange a symposium on Trichoptera. Letters from fellow workers suggested that now might be an appropriate time, and that a symposium should be held in Lunz. Today it is clear that large congresses are losing their value because of the difficulty of attending all relevant lectures and of finding colleagues. In consequence, small symposia for specialist groups are becoming increasingly important. As I felt that the success of such a symposium must depend on the suggestions from its potential members, I sent out in April 1973, together with a first circular, a questionnaire, asking for opinions on time and length of the symposium, numbers of participants, types of communication, interests within Trichopterology, and lan• guages which should be used. The majority of answers suggested the following: The number of participants should be between 20 and 50, and the duration, excluding excursions, should be three to five days; main interests were in ecology and systematics, but there were also interests in physiology, behaviour, zoogeography, morphology, cytotaxonomy and evolution. There was a clear preference for local excursions. Languages should be English, French and German, with a preference for English. The symposium should consist of both formal papers and informal progress reports, with adequate time for discussion. The arrangements have therefore been based on these results. The symposium which was held in the rooms of the House 'Zellerhof', was opened on the morning of 16th September. Two and a half days were devoted to lectures, and two half days to discussions. On the second day, a collecting excursion was made to the nearby valley of Seetal. In the evening of the first day the Mayor of Lunz, Mr. ENGELBERT HAGER, gave a reception in the rooms of the restaurant on the shore of the lake, and in the evening of the third day the four hundred years old building Amonhaus which is the Town Hall and local museum, was visited under the conduction of the well-known writer and historian, Mrs. ELISABETH KRAUS-KASSEGG. In the discussions the following topics were considered under the presidency of G. MARLIER: 1. Better information (Journal, Newsletter, etc.). It was proposed to start a Newsletter which would contain such useful information as: addresses of research workers, offers and demands for material and collaboration, etc. MALICKY would compile such a Newsletter, with the assistance of CRICHTON, at least for the first two numbers. If interest were maintained a continuation was possible and would be discussed later. The following workers were willing to act as correspondents for their countries and to help with the distribution of the Newsletter: BOTOSANEANU (Romania), BOURNAUD (France), CRICHTON (Great Britain), FLINT (USA, South America), FLORIN (Switzerland), HIGLER (Netherlands), KUMANSKI (Bulgaria), MALICKY (Austria, Greece), MARINKOVIC (Yugoslavia), MARLIER (Belgium), MORETTI (Italy), NEBOISS (Australia, New Zealand), NIELSEN (Denmark), NovAK

IX (Czechoslovakia), SOLEM (Norway), SZCZESNY (Poland), TERRA (Portugal), UJHELYI (Hungary), WIGGINS (Canada), ZINTL (German Federal Republic). A journal for Trichopterology seemed not realizable because of insurmountable economic problems. 2. Continuation of the Trichopterorum Catalogus. HIGLER reported on his efforts in this respect and asked for collaboration. 3. Identification literature, mainly for larvae. Identification works for Trichoptera larvae were urgently needed. MALICKY explained, in the absence of ILLIES, the project 'Limnofauna Mundi' which would consist of a number of books, edited by ILLIES, for the identification of all stages of aquatic animals of the world, but only, in principle, to genera. MALICKY asked for collaboration in the larval part of the Trichoptera volume; the adults would be treated by himself. Several objections were made to this type of identification book, mainly in connection with the continuing lack of information and the very different levels of knowledge in different regions of the world. It was agreed that it would be preferable to establish separate regional keys instead of a global one, but they could be collected later into a single volume. HILEY reported on keys for larvae in Britain on which he was working, MARLIER reported about a similar project for Africa. 4. Problems concerning the extinction of species and populations. Several exam• ples were reported from various countries. It seemed that the danger of extinction of single species of caddis-flies was not important except perhaps for endemics of small islands. Conservation of biotopes was much more important. BOTOSANEANU agreed to prepare a note concerning this point for the Newsletter. 5. Faunistics. Information was presented about the European Invertebrate Survey and the Sixth International Symposium on Entomofaunistics in Central Europe which would be held in Lunz am See in September 1975. 6. Speciation and evolution. BOTOSANEANU proposed collaboration on an interna• tional basis for the study of Wonnaldia occipitalis (Philopotamidae) which is one of the most promising objects for this kind of work. 7. Continuation of meetings, establishing a permanent organization committee. The continuation of this kind of meetings was agreed to CRICHTON would organize the next symposium in Reading (Great Britain) in 1977 to fit in with the SIL congress in Denmark. BOTOSANEANU, MALICKY and WIGGINS would assist in the organization. The other proposal, to hold it in or near Washington D.C., in connection with the International Congress of Entomology in 1976, was rejected because of the financial problems of attendance for many workers. Finally, I wish to express my gratitude to all those who helped in the organization of the meeting: above all, the Mayor of Lunz, Mr. HAGER, and his family; Dr. CRICHTON who helped with the correction of English texts; my wife; my father, Mr. J. T. MALICKY; and my collaborators of the Lunz Biological Station of the Austrian Academy of Sciences. I did my best to organize the meeting, but it was really made by the participants from 17 countries, by their presence, by presiding at sessions, by presenting papers, and by contributing to the discussions. My thanks are due to them all.

Lunz, March 1975 HANS MALICKY List of participants

ANTON ADLMANNSEDER, Dr., SchloBberg 8, A - 4910 Ried, Austria. RUTH M. BADCOCK, Miss, Dept. of Biology, University of Keele, Keele, Staffs. ST5 5BG, England. LAZARE BOTOSANEANU, Dr., Institut de Speologie, c.P. 2021, R - 7 Bucuresti 12, Romania. MICHEL BOURNAUD, Dr., Biologie animale et Zoologie, Universite Lyon I, 43 bd.du 11 novembre, F - 69621 ViIIeurbanne, France. YVETTE BOUVET, Biologie animale et Zoologie, Universite Lyon I, 43 bd.du 11 novembre, F - 69621 ViIIeurbanne, France. FERNANDA CiANFICCONI, Prof., Istituto di Zoologia, Universita, Via EIce di Sotto, 1-06100 Perugia, Italy. MICHAEL IAN CRICHTON, Dr., Dept. of Zoology, University of Reading, Whiteknights, Reading RG6 2AJ, England. CHRISTIAN DENIS, Dr., Laboratoire de Biologie animaIe, Premier Cycle, Faculte des Sciences, B.P. 25 A. F - 35031 Rennes, France. MICHAEL FEY, Ruhr-Universitat, Spezielle Zoo logie, D-463 Bochum, Germany. OLIVER S. FLINT, Jr., Dr., Dept. of Entomology, Smithsonian Institution, Washington D.C. 20560, U.S.A. JANETT FLORIN, Dr., HaldenstraBe 2a, CH - 9302 Kronbiihl, Switzerland. MICHAEL H. HANSELL, Dr., Dept. of Zoology, University of Glasgow, Glasgow, Scotland. HIGLER, BERT, Drs., Spotvogellaan 12, P.O. Box 184, BiIthoven, The Netherlands. PETER DAVID HILEY, Dr., Yorkshire Water Authority, Olympia House, Gelderd Road, Leeds 12, England. JOACHIM ILLIES, Prof., FluBstation des Max Planck-Instituts fUr Limnologie, Postfach 260, D - 6407 Schlitz, Germany. NEVILLE V. JONES, Dr., Zoology Dept., The University, Hull HU6 7RX, England. JOSEPH LE LANNIC, Lab. de Biologie animale 1. cycle, FacuIte de Sciences, Avenue General Leclerc, F - 35031 Rennes, France. DENISE LHONORE, Madame, Dr., Lab. d'Histophysiologie des Insectes, 12 rue Cuvier, F- 75005 Paris, France. HANS MALICKY, Dr., Biologische Station Lunz, A - 3293 Lunz, Austria. MARA MARINKOVIC-GOSPODNETIC, Prof., Prirodno-matematicki fakultet, YU -71000 Sarajevo, Yugoslavia. GEORGES MARLIER, Prof., Institut Royal des Sciences Naturelles, Rue Vautier 3 I, B-1 O~O Bruxelles, Belgium. GIAMPAOLO MORETTI, Prof., Istituto di Zoologia, Via EIce di Sotto, 1-06100 Perugia, Italy. JOHN C. MORSE, Dr., Dept. of Entomology and Economic Zoology, Clemson University, Clemson, South Carolina 29631, U.S.A. ARTURS NEBOISS, MSc., National Museum of Victoria, 285-321 Russell Street, Melbourne 3000, Victoria, Australia. ANKER NIELSEN, Dr., Zoological Museum, Universitetsparken 15, DK - 2100 KlZSbenhavn, Denmark. VINCENT H. RESH, Dr., Dept. of Biology, Ball State University, Muncie, Indiana 41306, U.S.A. HERBERT H. Ross, Prof., Dept. of Entomology, University of Georgia, Athens, Georgia 30602, U.S.A. KEENAN SMART, Dept. of Zoology, University of Glasgow, Glasgow, Scotland. STAMFORD D. SMITH, Prof., Dept. of Biological Sciences, Central Washington State College, Ellensburg Wa. 98926, U.S.A. JOHN O. SOLEM, Dr., DKNVS Museet, ErI. Skakkesgt. 47, N -7000 Trondheim, Norway.

Xl BERNHARD STATZNER, Zoologisches Institut, HegewischstraBe 3, 0 - 23 Kiel, Germany. LUIZ SILVEIRA WHYTTON DA TERRA, Estac;ao Aquicola, Vila do Conde, Portugal. SANDOR UJHELYI, Dr., Boniros ter 3, H-1093 Budapest, Hungary. F. VAILLANT, Prof., Allee de Pont Croissant, F - 38330 Montbonnot-Saint Martin, France. MARIA ILLUMINATA VIGANO-TATICCHI, Dr., Istituto di Idrobiologia, 1-06063 Monte del Lago sui Trasimeno, Italy. JAMES B. WALLACE, Dr. Dept. of Entomology, University of Georgia, Athens, Georgia 30602, U.S.A. WILFRIED WICHARD, Dr., Institut fiir Cytologie und Mikromorphologie, Universitat Bonn, GartenstraBe 61a. 0-53 Bonn, Germany. GLENN B. WIGGINS, Prof., Dept. of Entomology and Invertebrate Zoology, Royal Ontario Museum, 100 Queen's Park, Toronto, Ontario M5S 2C6, Canada. HERIBERT ZINTL, 0 - 8172 Lenggries, Germany.

_ ...... ;...... ;.. .. 2

Xli 3

4 Photographs from the Symposium: 1. from left to right: B. HIGLER, H. H. Ross, N. V. JONES 2. from left to right: S. UJHELYI, Mrs. MORSE, J. C. MORSE, H. MALICKY. 3. L. BOTOSANEANU speaking. 4. from left to right: W. WICHARD, M. FEY, V. H. RESH. (Photographs by J. T. M~L1CKY).

Xlll Proc. of the First Int. Symp. on Trichoptera, 1974. Junk, The Hague

Observations on the Helicopsychidae (Trichoptera) of New Caledonia

HERBERT H. Ross

To date, 12 species of Helicopsyche have been identified from New Caledonia, including an apparent local endemic from the Isle of Pines. In the case of five of these the larvae and adults have been associated. These species fall into four groups, the most primitive of which has adults that resemble somewhat closely several species occurring in Australia and New Zealand, such as heacota MOSELY and iltona MOSELY. These two and all the New Caledonia species have the clasper divided into definite dorsal and posterior lobes. These two and the two apparently most primitive New Caledonia species have no crease or furrow separating the two lobes of the clasper; all the other New Caledonia species have a crease or furrow separating the two lobes and on this basis would appear to be a uniquely derived group. A study of the world fauna will be necessary to support this point. The larvae of all the New Caledonia species construct basically a snail-like coiled case, but have three distinct patterns. Larvae of the primitive group 1 construct a smooth case of sand grains and with a low spire. In texture the case is much like the North American borealis, but the latter has a high spire representing the early narrow coils (Ross, 1944, Fig. 906). Larvae of groups 2 and 3 construct their cases of much coarser materials (Fig. la-g), using essentially small stones and virtually no sand, even when the case is as small as 4 or 5 mm in diameter. Normally they have a high spire, but the coils are not well separated from each other, rather they tend to merge into a generally uniform conical structure canted to one side (Fig. 1a,d). In a few collections of large cases containing healthy larvae, the spire has been removed either by the occupant or by some other agent. An unusual case belonging in group 3 is large for the genus, up to 22 mm in diameter, and has a set of radially arranged large stones set around the periphery of the case and attached to the large bottom whorl (Fig. If,g). In these the spire is relatively low. I originally called these side attachments ballast stones, because it seemed that they must contribute greatly to the stability of the case. Others have suggested a function of cryptic protective resemblance. These cases have been found in only two collections, both containing only large individuals, perhaps suggesting that smaller larvae of the same species may not add peripheral stones to their cases.

1 a c

d

Fig. 1. Cases of New Caledonia Helicopsyche. (a-e) typical cases of groups 2 and 3; (f,g) "ballast-stone" case; (h- k) case of group 4. (a-i) drawn to same scale; (j-k) more enlarged. (a,d,g,h,j) lateral; (b,i) dorsal; (c,e,k) ventral. 2 Larvae of group 4 construct quite a different case. In groups 1, 2 and 3 the cases are rigid and relatively thick walled, taking considerable strength to tear them apart. In group 4 the case is constructed from extremely fine sand grains and minute bits of detritus; these pieces are sufficiently small that the cross fibers of salivary gland secretion are obvious. Some cases are composed of at least half secretion. The walls of the case are thin and fairly flexible, but remarkably tough. The cases are small, attaining a maximum greatest diameter of 4 mm. They are basically a typical coiled case but in addition have a peripheral flange that ,continues the contour of the case down and outward to the surface of the substrate (Fig. 1h-k). The spire is low and from dorsal or lateral view the whorls are inconspicuous. The result is a case that looks like a limpet. When I first saw these pale sandy cases I assumed that their larvae frequented only sandy areas of stream bottom, but Dr. G. F. EDMUNDS, JR., who has collected many of them, tells me this is not so. He found them wandering over large and small rocks in the main stream and over debris of all sorts in backwater situations.

Acknowledgements

Dr. F. STARMUHLNER (University of Vienna), Dr. and Mrs. W. L. PETERS and Dr. W. M. BECK, JR., (Florida State A. & M. University), and Dr. G. F. EDMUNDS, JR. (University of Utah) and Dr. J. D. HOLLOWAY (Surrey, England) collected most of the material on which this study was based. I am indeed especially grateful to these colleagues. Through a research grant to Dr. W. L. PETERS, the National Geographic Society supported the collection of much material used in this paper. I am also extremely grateful to the National Science Foundation (U.S.A.) for a research grant in support of this study.

Reference

Ross, H. H. (1944). The Caddis-flies, or Trichoptera of Illinois. Bull. III. Natur. His!. Surv. 23: 1-326.

Discussion

WIGGINS: It is noteworthy that within the radiation of Helicopsyche species that seems to have occurred on New Caledonia, the highly unique helical case of the larvae has undergone at least two of the same structural modifications that the more usual tube case has undergone in other families. These modifications include the addition of heavier stones around the peripery, presumably to serve as ballast or balance in strong currents; and a peripheral flange of fine sand grains serving, it may be supposed, as a protective screen for grazing larvae. This seems to illustrate that specialization for certain food resources can be achieved by similar means even when the starting points are as different as a straight tube and a helix.

3 Proc. of the First Int. Symp. on Trichoptera, 1974, Junk, The Hague

A Progress report on the Phylogeny of Rhyacophila larvae

STAMFORD D. SMITH

This study will eventually encompass an evaluation of the phylogeny of all Rhyacophila larvae, but for the first two years emphasis has been placed on nearctic species groups. Larval and pupal (metamorphotype) specimens are now available for all major nearctic groups; missing are representatives of a few rare and isolated species from both eastern and western North America. When trying to establish main lines of developmerit within the genus, I have run into a great deal of difficulty. For example: structures such as gills may vary or be present in one group and absent in apparently closely related groups, although gills do not present the problems in nearctic species that they do in eurasian groups. The apico-Iateral spur of the anal prolegs does cause a great deal of difficulty, however. I am convinced that this structure has developed at least twice within the genus. The degree or extent to which characters only appear in later or last instars may be further complicating an evaluation of the phylogeny, but has opened up some interesting speculation upon which I have just started. Finally, cursory examination of larvae from both nearctic and pale arctic areas indicates that we may be dealing with three or four genera, however, most of the new genera will probably be very small and the bulk of the known species (3-4 hundred) will remain in the Rhyacophila s.s. and most certainly this genus will still be holarctic.

Discussion NIELSEN: The sclerite on the 10th sternum (the ventral site of the anal proleg) serves as the insertion for the flexor of the anal claw. I think that the nubila -fasciata group is the most advanced within the genus Rhyacophila. In this group the process on the basal sclerite of the anal proleg is very well developed. In my 1942 paper I have given an explanation of its function. ILUES: I understood that all larvae need two or three years for the development. SMITH: No. Mostly not. This is only the question in this one complex (Rhyacophila coloradensis) . ILUES: One species occurs in my greenhouses at different times of €merging. Do you think that the late summer generation is the direct offspring of the early summer generation? That would involve a development of only some months. If not so, we had to face the idea that there are two generations independently living along each other throughout the year. 5 SMITH: This is strongly indicated in some of the western North American complexes. At least in one situation it proved to be a case of cryptic species. I noticed definitely not one species with two generations a year. FLINT: Do you think it possible for some reason, a fraction of a generation emerges at one time of the year ~nd the other fraction at another time of year? Perhaps temperature might intervene to prevent complete emergence in the fall, or to delay development of some of the larvae in certain instars during the winter? SMITH: I don't know. NEBOISS: Does the number of larval instars vary? ILLIES: It would not explain the very decided gap between the generations. SMITH: I can say only about the Nearctic species that there may be a cryptic species, otherwise I don't know. CRICHTON: In Britain Rhyacophila dorsalis is normally univoltine but ELLIOT (1968) has shown that some individuals may spend a second winter as larvae. He records that there are five instars. VAILLANT: In the french Alps, Rhyacophila vulgaris develops at high altitude in 3 years, at lower altitude in 2 years and down in the valley in only one year; according to H. DECAMPS, R. intermedia does the same in the Pyrenees. Ross: I am sure this is not demonstrated in Trichoptera but in some other insects which have been studied carefully there are diapausing and non-diapausing generations in which these types are controlled by very simple genes. As a result, the population may have a polymorphic condition for this diapausing, such that every year a certain proportion diapauses and a certain proportion will not. This often leads to a difference in timing of generation appearance. WIGGINS: Are there, along with characters of the adults, larval characters that support subdivision of Rhyacophila into smaller groups? SMITH: I believe so, yes, but not necessarily destruction of the main part of the genus as it now stands however. WIGGINS: Considering the range of morphological types in Rhyacophila, is the general facies of the larvae, in so far as they are available for study, consistent with the smaller groupings that you are considering? SMITH: I think so, yes. HANSELL: I have a point which may be worth mentioning in relation to diapausing and non-diapausing of individuals within the same species. In Lepidostomq hirtum in Scotland the eggs are laid in July and the majority of individuals grow very quickly to reach the 5th'? instar by November-December. The 3rd instar individuals are very readily recognized since their houses are part sand grain and part leaf. So the main peak for 3rd instar will be seen in November but they have a subsiduary peak in about March suggesting that part of the population has delayed growth in some form followed by rapid growth to allow common emergence of the whole population in the summer.

6 Proc. of the First Int. Symp. on Trichoptera, 1974, Junk, The Hague

Contributions to the systematics of the caddis-fly family Limnephilidae (Trichoptera). III: The genus Goereilla.

GLENN B. WIGGINS

Abstract Goereilla baumanni DENNING, a caddis-fly of unusual phylogenetic significance, is recorded only from Montana; morphological data for the larva, pupa and female are presented for the first time. Because several derivative characters diagnostic for the Goerinae generally are represented in G. baumanni in a primitive condition, it is proposed that G. baumanni is the most primitive living member of the Goerinae now known. The species is considered to be a phylogenetic relict, together with Lepania case ada, another unusually primitive member of the Goerinae. The possibility is raised that the exceptional habitat of these species, wet, black muck of spring seepage areas, may be responsible for their continued existence as phylogenetic relicts.

Introduction

The Limnephilidae, one of the largest and most widely distributed families in the Trichoptera, have penetrated a wider range of aquatic habitats than any other in the order. More than 800 species are now known, and there are certainly many species still to be discovered. The family constitutes approximately one-third of the caddis• fly genera occurring in North America and Europe. Accepting the view that the genus represents the basic ecological unit (ILLIES, 1970, interalia), a premise that seems generally applicable to the Trichoptera, then for a relatively small order of insects this is indeed a noteworthy concentration of ecological diversification within a single family. Furthermore, the marked evolutionary success of the Limnephilidae is clearly a phenomenon of temperate latitudes. Ecologically, the principal role of larvae belonging to the Limnephilidae is reduction of plant materials to small particles. Vascular plant tissues are ingested by larvae of most species, and to a large extent these are materials of allochthonous origin. It is now clear (HYNES, 1970) that the primary source of energy for the biota of running waters is this allochthonous vascular plant tissue. Autochthonous algae are eaten as well, and in some groups larvae have scraper-edged mandibles that appear to be specialized for grazing upon periphyton (Fig. 15b). Knowledge about the important ecological role of the family is hampered by several problems in their systematics - arising for the most part from the larval stages.

7 At the alpha level, identification, reliable diagnostic characters for separating larvae at specific and generic levels are difficult to find, largely because of the relatively small number of larval associations available and because the range of variation in many of the characters is not sufficiently understood. Even in Europe and North America where there is a longer history of systematic work on larvae than elsewhere these problems are far from being solved. At the beta level, the fine studies by SCHMID (1955, etc.) have provided the first single classification for the world fauna of the Limnephilidae - an important achieve• ment. But this classification is based very largely on data from adults, and much work remains to be done in testing and augmenting its predictive value for larval-stages. Recent discoveries of larval stages in several North American genera indicate that there are discordances with the present classification still to be resolved (WIGGINS, 1973c, in press). Our understanding of phylogeny of groups within the family, the gamma level of systematics, has not kept pace with other advances. Although important contribu• tions have been made by NIELSEN (1943) and FLINT (196O), larval morphology has not been fully exploited for phylogenetic data, especially in some of the anomalous North American genera. A phylogeny for the family, well founded on comparative morphology of all stages, would sharpen insight into the evolution of ecological diversity within the Limnephilidae. Initial exploration along these lines for species living in temporary pools has been made (WIGGINS, 1973a).

Current Studies

The objectives of my studies in the systematics of the Limnephilidae are to discover more of the basic taxonomic data about the larvae which will enable ecologists to work more precisely with them at generic and species levels, and to use data from this source and from others to gain a better understanding of the systematics of the Limnephilidae in both classification and phylogeny. Initial emphasis has been on faunal exploration in North America where a great deal of systematic synthesis is to be done, even in establishing larva-pupa-adult associations at the generic level. Phylogenetic analysis of larval morphology has revealed that certain genera in western North America, e.g. Moselyana, Imania and Manophylax, represent aberrant phyletic lineages discordant with the subfamilial classification that accommodates the world fauna generally (WIGGINS, 1973c). Work on limits and definitions of the limnephilid subfamilies continues with comparable analyses for characters of adults. Because of the number of aberrant forms in western North America, the area must be regarded as one of unusual interest in the study of limnephilid systematics and evolution. Current work concerns phylogenetic relationships between the Goeridae (auct.) and the Limnephilidae (s.s.). With discovery of the larva of Lepania cascada (WIGGINS, 1973b), the definition of the Goeridae was enlarged to encompass forms with ocelli in the adults and toothed mandibles in the larvae - both plesiomorph

8 conditions of the characters involved. The consequent reduction of the taxonomic gap between the Goeridae and Limnephilidae (s.s.) supports the classification advo• cated by NIELSEN (1943) and FLINT (1960), viz. subfamily Goerinae of the Lim• nephilidae (s.l.). This classification is accordingly being followed in current studies, representing the working hypothesis that evolution of the Limnephilidae and Goeridae is so closely interconnected that the two taxa should be united in a single family, at least until there is firm evidence to support separation of the Goeridae as a family group. Because of toothed mandibles (Fig. 14b) and short basal spurs on the tarsal claws (Fig. 14c) in the larva, and ocelli in the adults (Fig. 14d), Lepania cascada was proposed as an unusually primitive representative of the Goerinae (WIGGINS, 1973b). Although adults of another goerine species, Goereilla baumanni (DENNING, 1971), were also known to have ocelli, lack of knowledge of immature stages prevented any further assessment of its phylogenetic relationship. In the course of my field work in Montana in August 1973, unusual goerine larvae were found at the type locality of G. baumanni. Mature larvae that had sealed themselves within their cases in August were brought back to our laboratory and held at approximately 4°C. These larvae developed to pharate adults by March 1974, and their identity as G. baumanni was confirmed. Because morphology of the larva shows several unusual features that are important in interpreting systematic relationships between the Limnephilidae (s.s.) and the Goerinae, it is appropriate to give here a summary of morphological and biological data for G. baumanni. Structural characters of the female of this species, not yet available in the literature, are outlined, and also those of the male to provide a complete diagnosis for this interesting species.

Data for Goereilla baumanni LARVA. The larva of this species is distinctive from all others and can be readily recognized by the enlargement of the mesepisternum into a rounded lobe bearing stout spines. Length of larva up to 9 mm. Most sclerotized parts reddish brown in colour. Head (Fig. 4) squarish in dorsal view, lateral margins parallel, surface coarsely pebbled around periphery, plate-like sculpturing on dorsum, secondary setae absent, antennae in small depressions, ventral apotome (gular sclerite, Fig. 9) with posterolateral margins convex. Labrum (Fig. 6) with anterolateral border unpigmented but sclerotized, dense row of setae in median notch. Mandibles (Fig. 5) as in Lepania with distinctly toothed cutting edges, each with mesal brush of barbed setae. Labium (Fig. 9) with submental sclerites triangular, not quadrate as in Lepania; each bearing single stout seta, not two as in Lepania; sclerite of palpiger angulate as in Lepania, not extending around base of labial palp; maxilla differing from Lepania in having only single lateral seta on sclerite of stipes; galea somewhat more lobate than in Lepania; maxillary palp with lateral fringe of setae, restricted to base of palp in Lepania. Pronotum (Figs. 1,4, l3a) not conspicuously enlarged ventrolaterally as in Lepania and other goerines (Figs. l4a, l5a), surface coarsely pebbled, mostly covered with long setae, anterola• teral margins with a few coarse teeth; prosternal horn well developed. Mesonotum with two pairs of scIerites on each side, presumably representing sal and sa3, sa2 scIerites fused across mid-dorsal line; mesopleuron (Figs. 1, 2, 13a) modified from normal condition with mesepister• num extended anterad as short, rounded lobe bearing stout spines, but not as prominent process of Lepania and other goerines (Figs. l4a, l5a); mesonotum lacking short transverse ridge between sal and sa2 scIerites characteristic of Lepania and other goerines (Figs. 14a,

9 ISa). Metanotum with each primary setal area represented by separate sclerite much as in Lepania. Legs with basal seta of tarsal claws much shorter than claw; hind legs distinctly shorter than middle legs. Abdomen (Fig. 1) with all three humps of segment I well developed; small setae between median hump and each lateral hump 165-184 (sample of 5 larvae); setae on venter I 79-97(5) and also a transverse, median ovoid sclerite, weakly defined and pigmented. Gills single, in final instar larvae two dorsal and two ventral gills present on each side of segments II-VII, single dorsal and ventral gill on VIII. Lateral line with sparse, small setae; all segments lacking sclerotized ventral rings. Dorsal sclerite of segment IX wide and short, bearing many setae all of approximately same length. Anal prolegs (Figs. 7, 8) with tiny spines on dorsum of each lateral sclerite; basal tuft of three stout setae arising from small, transverse sclerite; claw of anal proleg with prominent accessory tooth. Larval case (Fig. 3) of small rock pieces with some wood fragments, a tapering curved cylinder much as in Lepania, but posterior opening circular and with crenulate margin rather than crescentic as in Lepania. Length of larval case up to 11 mm.

PUPA. The pupa is more similar to Goeracea and Goerita than to Lepania, but distinguished from all these genera by the shape of the anal processes and their three long subapical setae. Length of pupa up to 9.5 mm. Head (Fig. 10) with cutting edges of mandibles convex proximal to apex, fine teeth along concave margin before pointed apex; labrum with group of five stout setae on each side near base, their apices twisted; two setae on each side at base of labrum. Abdomen (Fig. 12) with sclerotized plates as illustrated; gills located as in larva, reduced to short stubs; anal processes short and stout, surface pebbled, apices curved dorsad and pointed; apex of each process with three stout setae, approximately twice as long as process, arising before apex from mesal edge. Pupal case with anterior closure (Fig. lla) incorporating several pieces of wood, a few small holes between the pieces; posterior closure (Fig. lIb) apparently little altered from larval case. Length of pupal case up to 11.5 mm.

ADULT. Adults of G. baumanni can be distinguished from all other species by characters of the male and female genitalia. Length of fore wing: 0, c;> 7-8 mm. Sclerotized areas black to dark brown. Head (Fig. 19) with two pairs of rounded warts between median ocellus and lateral ocelli, elongate posterior wart behind each lateral ocellus, warts on front of head as in Fig. 20; maxillary palpi of male (Fig. 20) three-segmented as in Lepania, but not curved dorsad before face, terminal segment approximately same length as middle segment, slightly enlarged apically but segments not modified as in other genera such as Goerita (WIGGINS, 1973b, Fig. 24). Thorax (Fig. 19) as in Lepania; pronotum with pair of rounded warts; mesoscutum with pair of small, elongate warts; mesoscutellum with single elongate wart. Spurs 1, 2, 2. Wing venation (Fig. 18) similar in the sexes; hind wing with frenulum of approximately six stout setae, followed by series of short, stout marginal setae. Male genitalia (Fig. 16). Distinguished from all other species by the shape of the praeanal appendages and the pair of pointed processes of segment X. Segment IX narrowed dorsally. Claspers two-segmented, basal segment stout, terminal segment simple and finger-like. Segment X with pair of large, lobate praeanal appendages excavate posteriorly; ventral to these a pair of straight, elongate processes, each pointed apically, and each apparently combining a dorsal element (internal branch, segment X?) and a ventrolateral element (external branch, segment X?) fused along lateral ridge. Phallus with invaginated endotheca bearing patch of spines. Female genitalia (Fig. 17). Distinguished from all other species by the simple ventral lip of the genital opening.

10 ~.''::'''.~

W8

Figs. 1-8. Larva of Goereilla baumanni. (1) larva, lateral X xI3.5; (2) mesopleuron, lateral; (3) larval case with posterior opening X x9; (4) head and thorax. dorsal; (5) left mandible, ventral, with seta from tuft enlarged; (6) labrum, dorsal; (7) segment IX with anal prolegs, dorsal; (8) claw of right anal proleg, lateral.

11 Figs. 9-12. Goereilla baumanni. (9) larval maxillae and labium with gular sclerite, ventral; (10) pupal head, frontal; (1la) anterior closure of pupal case; (lIb) posterior end of pupal case; (12a) pupal abdomen with sclerotized plates enlarged, dorsal; (12b) anal processes, dorsal; (12c) anal processes, lateral. -

12 Goereilla 13d

pP

Lepania

Goeracea 15d

Figs. 13-15. (13) Goereilla baumanni; (14) Lepania cascada; (15) Goeracea genota. (a) hhval head and thorax, lateral; (b) right larval mandible, ventral; (c) larval tarsal claw, lateral; (d) head of adult, dorsal. 13 Similar to Lepania in having recognizable venter to segment IX, separate from VIII laterally but joined to VIII ventrally and terminating in transverse, simple, rolled lip that forms ventral edge to genital opening. Posterodorsal to genital opening is partially sclerotized supragenital plate subtending anal opening. Terga of IX and X apparently fused along transverse sulcus, X terminating in two rounded lobes, wart-like prominence at the base of each lobe.

MATERIAL STUDIED. MONTANA: Missoula Co., spring entering Butler Creek at Snow Bowl ski area nr. Missoula, 17-18 August 1973, G. B. WIGGINS, many larvae, prepupae (14 reared); same locality, 26 May 1974, D. S. and L. M. POTIER, many 00, 2';>';>. Flathead Co., spring at mile 11, east side Hungry Horse Reservoir nr. Martin City, 6 Aug. 1973, G. B. WIGGINS, 3 larvae.

BIOLOGY. Larvae were collected from the water-saturated organic muck at the edge of a spring seepage area; the emergent spring run, 15-30 cm wide, flowed down the side of a steep valley for approximately 20 m to join Butler Creek. No larvae were found in the spring run or in the creek.

16c 16d

17a 17b

Figs. 16-17. Goereilla baumanni genitalia. (16a) male, lateral; (16b) male, dorsal; (16c) male, ventral; (16d) male, phallus, lateral; (17a) female, lateral; (17b) female, ventral.

14 Figs. 18-20. Goereilla baumanni adult. (18) wing venation of male; (19) head and thorax of male, dorsal; (20) head of male, frontal.

Head widths of larvae collected in August comprised two groups: 0.85-0.90 mm (48 larvae, 5th and final instar), and 0.60-0.65 mm. (4 larvae, probably 4th instar); a few larvae were sealed in their cases, evidently in preparation for pupation. These latter larvae and a number of the active final instars were maintained alive in a laboratory refrigerator at approximately 4 0c. Some reached the pupal stage by December when they died; others were pharate adults by March. Since normal emergence of adults occurs in May, these observations indicate that the insects are sealed up in pupal cases as final-instar larvae, pupae, and finally pharate adults for approximately nine months. In few, if any, other caddis-flies is there known to be as long a period between the active and feeding final-instar larva and the adult. Absence of younger instars from the series suggests a life cycle of one year.

Discussion concerning Goereilla baumanni

Even though the larva of G. baumanni lacks most of the apomorph characters diagnostic for other goerines, and even though the adults possess ocelli, the wing venation and genitalic structure support the original classification of the species by DENNING (1971). Within the context of that general position in the classification, several larval characters of Goereilla baumanni are particularly significant. a. The cutting edge of each mandible has several pointed teeth, and the basal seta of all tarsal claws is much shorter than the claw. These characters, both plesiomorph for the Lim'1ephilidae (WIGGINS, 1973c), are shared with Lepania cascada (Figs. 13b,c; 14b,c). All other members of the Goerinae now known exhibit the apomorph condition-cutting edge of the mandibles with separate points modified to form a 15 continuous scraping edge (Fig. 15b), and the long basal seta of all tarsal claws reaching almost to the tip of the claw (Fig. 15c). b. The pronotum in G. baumanni (Fig. 13a) is similar to the plesiomorph condition in the Limnephilidae generally, and lacks the apomorph ventrolateral enlargement present in other Goerinae (Figs. 14a, 15a). c. The mesonotum in G. baumanni (Figs. 4, l3a) lacks the transverse ridge between sal and sa2, an apomorph feature that characterizes other members of the Goerinae (Figs. 14a, 15a). d. The mesepisternum (Figs. 1, 2, 13a) is produced anterad as a short, rounded lobe bearing stout spines. Although this clearly is an apomorph development from the normal condition of the mesepisternum in the Limnephilidae and related families, it is far less developed than the highly apomorph anterior extension of the mesepister• num in other members of the Goerinae (Figs. 14a, 15a).

Thus G. baumanni is the only goerine now known in which the larva exhibits a plesiomorph condition of the derivative characters (a-d above) diagnostic for the group. This condition could be a result of regressive evolution from the typical goerine condition, not an unreasonable possibility for the structure of the mesepister• num (d above). But the concomitant requirement for regression in all these charac• ters seems less likely. It is reasonable, then, to interpret this evidence to mean that G. baumanni is in reality a plesiomorph goerine, and a representative of the most primitive line yet known. Previous comments concerning the unusual phylogentic significance of Lepania cascada (WIGGINS, 1973b) are therefore to be extended, and with added emphasis, to apply to G. baumanni as well. In accordance with this interpretation, these two species may rightly be considered phylogenetic relicts (sensu DARLINGTON, 1957). For the present, Goereilla baumanni seems best assigned to the tribe Lepaniini together with L. cascada because of the primitive characters retained in both. There is little real similarity between them in terms of derivative characters (WIGGINS, 1973b), but I think it better to defer creation of another monotypic tribe until a review of the Goerinae as a whole can be undertaken. The habitat in which larvae of G. baumanni were discovered is the wet, black organic muck of a spring seepage area; samples of the clear sand and gravel substrate of the spring run itself yielded some empty cases, but none with larvae. These sites are unusual habitats for caddis-fly larvae, but seem all the more significant because this is precisely the habitat in which we found larvae of L. cascada and also Moselyana comosa, in Oregon (WIGGINS, 1973b,c). The larval habitat of the two goerine species, along with their toothed mandibles (Figs. l3b, 14b), suggest that they are detrital feeders. Gut contents of larvae of G. baumanni (3 larvae) and of L. cascada (3) were largely vascular plant pieces with smaller proportions of fine particles and algae, but with very few mineral fragments. By comparison, guts of larvae of Goeracea genota (3) and of Goera sp. (3) contained a higher proportion of fine particles than vascular plant pieces, but contained a

16 considerable amount of mineral fragments. Larvae of both Goeracea and Goera inhabit open waters of streams and have mandibles modified for scraping rock surfaces to which sand particles adhere (Fig. ISb), a feeding behaviour consistent with the presence of mineral fragments. The question arises, then, whether the highly specialized larval habitats of species such as G. baumanni and L. cascada are in some way responsible for their continued existence as phylogenetic relicts. Competition and predation are evidently low in organic muck, for apart from earthworms and an occasional dipteran larva, our collections in these sites yielded few other animals. The vascular plant pieces ingested by the caddis larvae are abundant because there is a rich growth of mosses and herbaceous plants on the surface of the muck; yet our collections in these sites have never yielded an abundance of the caddis-flies. In conclusion, I should like to open the question for discussion: Are there species of caddis-flies in the European fauna which are characteristic of the wet organic muck of spring seepage areas?

Acknowledgments

Field work in Montana in 1973 was supported by the National Research Council of Canada (Grant AS707). I am indebted to Dr. JOHN F. TIBBS, Director of the University of Montana Biological Station for accommodation and working facilities at the Station during my period there as a Visiting Investigator. Mr. DAVID POTTER guided me to collecting localities in August and obtained a series of adults of G. baumanni the following spring. Mr. TOSHIO YAMAMOTO of our Department was responsible for rearing larvae of Goereilla baumanni to the adult stage. Larval gut contents were identified by Dr. F. BAERLOCHER. Dr. D. G. DENNING loaned me adult specimens of G. baumanni. The illustrations were prepared by Mr. ANKER ODUM, scientific illustrator, Department of Entomology and Invertebrate Zoology, R.O.M.

References

DARLINGTON, P. J. 1957. Zoogeography: the geographical distribution of animals. John Wiley & Sons, 675 pp. DENNING, D. G. 1971. A new genus and new species of Trichoptera. Pan-Pacif. Ent., 47(3): 202-210. FLINT, O. S. JR. 1960. Taxonomy and biology of nearctic limnephilid larvae (Trichoptera), with special reference to species in eastern United States. Entomologica Am., n.s., 40: 1-117. HYNES, H. B. N. 1970. The ecology of running waters. Univ. Toronto Press, 555 pp. ILLIES, J. 1970. Die Gattung als okologische Grundeinheit. Faunistisch-Okologische Mit• teilungen, 3(11/12): 369-372. NIELSEN, A. 1943. Trichopterologische Notizen. Vidensk. Meddr. Dansk. Naturh. Foren., 107: 105-120. SCHMID, F. 1955. Contribution 11 I'etude des Limnophilidae (Trichoptera). Mitt. Schweiz. Ent. Ges., vol. 28, 245 pp. WIGGINS, G. B. 1973a. A contribution to the biology of caddis-flies (Trichoptera) in temporary pools. Life Sci. Contr., R. Ont. Mus., no. 88: 1-28:

17 WIGGINS, G. B. 1973b. New systematic data concerning the North American caddis-fly genera Lepania, Goeracea and Goerita (Trichoptera: Limnephilidae). Life Sci. Contr., Onto Mus., no. 91: 1-33. --. 1973c. Contributions to the systematics of the caddis-fly family Limnephilidae (Trichopt• era). I. Life Sci. Contr., R. Ont. Mus., no. 94: 1-32. -- (in press). Contributions to the systematics of the caddis-fly family Limnephilidae (Trichoptera). II. Can. Ent. Discussion ILUES: In the European fauna, in the limnephilids it is just the genus Parachiona which is completely restricted to springs, and as far as I know no larvae are known. NIELSEN: I have described it. ILUES: Have you? Pardon. NIELSEN: It belongs to what VAILLANT had called the madicolous fauna. HILEY: How do these species living in organic ooze manage to obtain material for their mineral cases? WIGGINS: This is a difficult question to answer. Because of the method of collection - handfuls of wet soil are washed in running water to remove muddy material- we do not know at what level the larvae actually live in relation to mineral substrate. But there is always some mineral left in every sample after the organic material is washed out, and I should think particles of sand-grain size are probably suspended throughout the wet soil. Certainly the larvae of Goereilla find them and are evidently highly selective in doing so. HILEY: What proportion of the total mass of ooze is mineral? WIGGINS: It is a very small one. VAILLANT: I have to ask Dr. WIGGINS about his opinions on the relations of the Lepidos• tomatidae to the Limnephilidae. WIGGINS: My opinions on this are not yet very advanced. I am impressed by the morphologi• cal similarities between the two, and I am impressed with the ecological similarities between the two groups. I just have not yet assessed the data for the Lepidostomatidae phylogenetically and thus cannot yet make any comment. NIELSEN: Your opinions seem very convincing, facts are facts. It leaves, however, the Apataniinae somewhat 'floating in the air'. Some people consider this subfamily as an intermediate between Limnephilinae and Goerinae. I don't know whether the European genus Lithax is represented in North America. The Lithax larva is a typical Goerinae larva, but its house is almost like that of Apatania. WIGGINS: I hope to be able to offer views on the relationships of Apatania and the Apataniinae, in due course. My comments here dwelt on the Goerinae because we have discovered phylogenetic data of unusual significance in larvae of some of the North American goerine species. I am aware that there are indeed difficult questions concerning other subfamilies that have yet to be resolved. NIELSEN: Do these primitive larvae feed like the other goerids by eating the microflora of the stones0 WIGGINS: Larvae of the limnephilid genus Neophylax, which is well represented in North America, graze upon the microflora of rocks, too.

18 HILEY: The Goereilla larvae you have described are similar in some respects to certain limnephilids, and in other respects to the Goeridae, which I consider'a part of the sericos• tomatid group. I think one could find, on the basis of larval characters, a good chain of similarities from Goera and Silo through various genera to limnephilids like Drusus and Apatania. WIGGINS: I shall make two comments. One of the reasons for my coming to this Symposium is to have my first opportunity to observe and collect European Apataniinae in their natural habitats, and so to learn more about this group. The non-toothed or scraper mandibles seem clearly to be associated with the grazing of periphyton and other micro-organisms: I suspect that the long basal spur on the tarsal claw is also related to this kind of feeding because it is usually present in larvae that have non-toothed mandibles. HILEY: What about the secondary setae on the first femur? Could they also be related to this habit? WIGGINS: That is a possibility. Most genera in the subfamily Dicosmoecinae have multiple setae on the femora, but they have toothed mandibles as do most larvae in the Limnephilidae. My impression is that we do not yet have enough detailed information about food habits to effectively analyse the significance of morphological features related to feeding. NIELSEN: A very short comment. You asked on European caddis-fly larvae living in certain organic ooze. These are the Sericostoma larvae.

19 Prac. af the First Int. Symp. an Trichaptera, 1974, Junk, The Hague

The identification of British Limnephilid and Sericostomatid (s.l.) larvae

PETER D. HILEY

The main reason for giving this paper is to advertise the two keys to larvae that I have made, and to explain some problems which remain. I cannot continue the research, due to lack of time, and therefore hope that others will take up the study and perhaps resolve these problems. . Previous keys have suffered from inadequate rearing methods. In general the following method has been used: From a collection of larvae presumed to be monospecific several were preserved and the remainder reared to adults for identifi• cation. In groups such as Limnephilidae where the larvae of many species are very similar this method is obviously open to error. In my rearing method (HILEY, 1969) the prepupa is removed from the pupal case and allowed to shed its larval skin outside the case. The pupa is then kept alive until ready to emerge, at which stage identification is usually possible. Thus for each larval skin, which is a reasonably complete record of the larva, one has the resultant pupa and false identification is not possible. By this method one can produce separate sets of accurately known larval skins for groups of species which are very similar as larvae. Variable species can be reared without selective bias as to colour, form, etc. I collected the larvae of 41 species of Limnephilidae in Britain and reared them in this way. Colleagues in Britain supplied larvae of two more species, and larvae of a further three species were sent from France by colleagues there. Information from the literature was used for eight more species, leaving the larva of only one species, Limnephilus fuscinervis, completely unknown. I also collected larvae of all nine species of the British sericostomatid-group, and reared eight of these. Since I had so much new material, I decided to make completely new keys based on characters deduced from my specimens. I attempted, wherever possible, to use positive characters and avoid those which overlapped in some instances. I obtained 50 skins of one species, and this set of material served as a type. By comparing all the other species with it, one by one, it was possible to deduce characters which would be of value in many parts of the key. This method produced characters which enabled me to divide the Limnephilidae into groups of up to ten species. It was then possible to subdivide each group by comparing its members with each other. Where possible, I studied at least twenty examples of each species in order to catalogue the

21 variability of the selected characters. In several cases it was obvious that the total variation of a character increased when specimens from new localities were ex• amined. There may, therefore, be some regional variations which I have not noted, and further study of these would be of value. I attempted to make the routes to all the species of Limnephilidae of similar length and as short as possible, but in some cases this meant making the routes to a few species a little longer. The key to the sericostomatids (HILEY, 1972 and addendum in prep.) separates every species. Such a complete separation was not possible in the Limnephilidae, where no positive characters were found for the separation of these five groups of species: (1) Halesus radiatus, H. digitatus, Hydatophylax infumatus; (2) Stenophylax permistus, S. vibex, S. lateralis, S. sequax; (3) Potamophylax latipennis, P. cingulatus, P. rotundipennis, Allogamus auricollis, Chaetopteryx villosa; (4) Limnephilus affinis, L. incisus; (5) Limnephilus lunatus, Mesophylax aspersus. Some information is given in the key on features which may aid identification, such as time of emergence, type of case, etc. I expect that the final separation of these groups will entail a considerable amount of work. I think it would be appropriate to explain some of the problems remaining, though these are treated more fully in the limnephilid key which should be published in 1976. Specimens of H. infumatus were not obtained, and since there is considerable disagreement between the descriptions of HANNA (1957) and LEPNEVA (1971), the only solution is to obtain larvae and rear them. I compared FROCHOT'S (1962) larvae of S. permistus with his illustrations and found several underestimates of setal counts on the nota. Because of the chance that HICKIN (1967) may also have omitted some setae on these areas in his illustration of S. vibex, I decided to wait until larvae were available before using a possible chaetotaxy character. Specimens of A. muliebris from Britain showed great variation from place to place, and some differed consider• ably from NIELSEN'S (1942) specimens which he collected in northern Jutland. Variation in the other two British Apatania species has added to the confusion. Only A. muliebris and A. wallengreni are separated in my key, but Messrs. J. O'CONNOR, I. D. WALLACE, Dr. E. J. WISE and myself intend to publish a special paper dealing with the problems in Apataniinae when we have examined more material. I shall welcome any comments, at any time, regarding these keys. References FROCHOT, B. 1962. La larve de Stenophylax permistus McL. (Trichoptera, Limnophilidae). Trav. Lab. Zool. Stn. aquic. Grimaldi fac. sci. Dijon, 42: 1-19. HANNA, H. M. 1957. The larva of Stenophylax infumatus McL. (Trichoptera: Limnephilidae). Entomologist's Gaz., 8: 218-22. HICKIN, N. E. 1967. Caddis larvae. Larvae of the British Trichoptera. Hutchinson, London. 476 pp.+xi. HILEY, P. D. 1969. A method of rearing Trichoptera larvae for taxonomic purposes. En• tomologist's mono Mag., 105: 278-9. HILEY, P. D. 1972. The taxonomy of the larvae of the British Sericostomatidae (Trichoptera). Entomologist'S Gaz., 23: 105-19. LEPNEVA, S. G. 1971. Fauna of the U.S.S.R. Trichoptera. Vol. II, No. II. Larvae and pupae of Integripalpia. Israel Program for Scientific Translations, Jerusalem. 638 pp.

22 NIELSEN, A. 1942. Uber die Entwicklung und Biologie der Trichopteren mit besonderer Beriicksichtigung der Quelltrichopteren Himmerlands. Arch. Hydrobiol. Suppl., 17: 255- 631.

Discussion

Ross: What genera do you have in the Lepidostomatidae? HILEY: Lepidostoma, Crunoecia, and Lasiocephala. WIGGINS: What setal characters did you use for separating genera? HILEY: suppose that some of the main characters are on the femora, for instance the secondary setae on the first femur are used for separating Apatania, Drusus and Ecclisopteryx from the rest of the single-gill group in Limnephilidae. In other instances the secondary setae of the second and third femora were useful. I have made counts of setae on the heads, nota and 'anal apparatus', but found few instances where the range of counts from closely similar species did not overlap. Characters involving the length of setae are problematic, since they get broken rather easily, though a character involving the diameter of the basal area may be possible. Mr.

WALLACE has taken up the study, and he has found some possible character~ which involve the setae of the abdomen. This was an area which received relatively little attention from me, and the results demonstrate how valuable a fresh approach can be. WIGGINS: Did you use the arrangements of the setae on the first abdominal segment? HILEY: There is a character of this area deduced by IAN WALLACE to aid in the separation of Mesophylax impunctatus from its close neighbours Limnephilus extricatus, L. fuscicornis and L. griseus, but it does not in fact appear in my key. WIGGINS: We have done work on setal characters of the first abdominal segment based on both setal number and setal arrangement. One can usually recognize the 3 primitive setal areas on both the dorsal and the ventral surfaces of this segment, and there are many modifications of the basic arrangement. And in certain genera there are weakly sclerotized areas around the base of the lateral humps. Have you used these characters? HILEY: My eyes were fixed only on the nota, limbs, gills, etc., and it appears that I could not see possible characters on the first abdominal segment. ILUES: The older authors put much emphasis on the presence or absence of these balance stems on the side of the case in Halesus for instance. Did you find the cases of your species distinguishing? HILEY: I found some features of the cases very useful, but many characters are variable, and most cases change with growth. Where I was in doubt I tried to make the larvae build a case different from the one it usually makes. For instance I found that Halesus could make a very similar case to Potamophylax, i.e. completely of stones. In the Stenophylax I could not find any characters of the cases which could be used for firm separation. In the Limnephilini group they will either build a straight or curved case, and this is a very consistent character which can be used. I found that it was not possible (with one exception!) for a curved case builder of this group to make a straight case, and vice versa. ZINTL: Is there reason to suspect that the larvae of Allogamus and Chaetopteryx make a style change on their cases through the instars? 23 HILEY: Yes. In all of the species of Limnephilidae. The cases used in the key are the final (5th) instar only. There are very great changes through the life of all the case building caddis-flies I studied. I think it is a very worthwhile study and hope that others will follow the fine example set by yourself and Dr. HANSELL. NEBOlSS: How can you recognize the 5th instar larva? HILEY: You can, if you have larvae near to pupation. The fat body in limnephilids does not really develop until the final instar. In earlier instars, and very young final instars, there is a characteristic translucence of the body, and some of the internal organs are visible. The highly developed fat body in the mature final instar causes a very characteristic colour. This can be white, orange, yellow, blue, green. In some species it is sex-linked, for example Limnephilus lunatus males are green, females yellow. In other species both green and yellow occur but are not sex-linked. The color and size of the fat body is of considerable value, especially in the field, but it is difficult to describe it. You really have to see it. Otherwise I give the length of the fully grown larva. Fully grown final instars may also have an empty gut, whereas the gut in earlier stages usually contains food, which is visible. BOURNAUD: Have you found differences between Micropterna and Stenophylax? HILEY: I have a tentative character to separate 'Micropterna' sequax and 'Micropterna' lateralis from Stenophylax permistus and S. vibex. The dorsal part of the case is slightly curved in Micropterna, and straight in Stenophylax. This is the only character I have seen. I have found none on the larvae. NIELSEN: I was very pleased to see brevipennis CURT. placed side by side with Limnephilus decipiens KoL., for there it belongs. Both as to morphology and ecology of the larva as well as to the physiology it is quite impossible to include brevipennis in the genus Anabolia. It has -like many other Limnephilus species - an imaginal summerdiapause, whereas the Anabolia species have a larval summerdiapause.

24 Proc. of the First In!. Symp. on Trichoptera, 1974, Junk, The Hague

Some Philopotamidae from France

F. VAILLANT

In Europe, there are only 6 species of the genus Philopotamus LEACH, and 4 of them occur in France; one, Ph. corsicanus McL., is endemic of Corsica. The other 3 seem to have different ecological requirements, for their territories overlap very little. Larvae of Ph. ludificatus McL. are extremely abundant in unpolluted torrents of the French Alps anq, under almost every big stone over which water flows, several can be found in early spring; they seem to be more numerous in limestone mountains than in igneous ones. I never found a single specimen of Ph. montanus (DONOVAN) in the Alps, but this species is, in the Massif Central, even more abundant than Ph. ludificatus is in the Alps; larvae are found on all kinds of rocks, but are rather rare on limestone; indeed this may be due to low elevation. As for Ph. variegatus McL., it seems to be rather rare everywhere in France, though specimens were collected in different parts of this country. I examined many male specimens of Ph. ludificatus from several sections of the French Alps and among them some collected not far from the mediterranean coast; their genital parts are all alike. The phallus of Ph. ludificatus has, on its ventral side, a long chitinous plate bent downward at its tip; above it, inside the phallus, is a long and almost straight spine and, in front of it, is a complex piece with two hooks directed downward and laterally and with a fork-shaped sklerite (which has been partly drawn on one side on Fig. 1). All these pieces were perfectly alike for all the specimens from the French Alps I examined (Figs. 1 and 2), but they are a little different for specimens from the Austrian Alps; Fig. 3 shows the complex piece of a specimen collected beside the stream flowing out of Lunzer Mittersee; the hooks of this piece are more stout and less curved than they are for Ph. ludificatus from western Alps and the fork-shaped sklerite is split on the medial line. Ph. montanus has at least 4 varieties in Europe; they differ by the number and the relative size of the spines of the phallus; these characters seem to be much more important than those that distinguish the two varieties of Ph. ludificatus. Another genus of the family Philopotamidae is Wormaldia McL.; the imagos are small brownish caddis-flies. Our knowledge is incomplete concerning certain species of Wormaldia, such as W. mediana McL. and W. subnigra McL., and we do not know whether they have or not several varieties. But it is likely that W. mediana has uniform characters 25 throughout its territory; according to H. DECAMPS, its larvae live in large rivers at low altitude, so that it has been able to spread its areal without any tendency towards endemism. w. triangulifera triangulifera McL. has also a wide range and its larvae live in rather large streams; the species is found in Spain, in the Massif Central and also near the Morvan low mountains, which are separate from the latter. Quite to the contrary W. occipitalis Pict. has many subspecies and varieties; this is probably due to geographical isolation, for the larvae of this species develop in small· shallow springs under stones, in moss or among decaying leaves, most always under a thin layer of water; they can withstand a madicolous environment and remain on rocks or stones over which water trickles; I never found any in streams or torrents. The varieties of W. occipitalis are endemic of a small territory; maybe some have their range restricted to one spring isolated from others. As it has been pointed out by L. BOTOSANEANU, in a population of W. occipitalis, characters vary inside of close limits, and between this population and another, there is usually a distinct gap and a discontinuous change of characters, especially of those of the genital parts. The different varieties of Worrnaldia can be distinguished only by characters of the wing venation and the number, size and shape of the spines in the phallus. In W. occipitalis subsp. occipitalis var. A, from the french Alps, there are, in the phallus, 6 large isolated spines, named A, B, C, D, E and F, the last two being close to the end of the phallus, and 4 clusters of smaller and more slender spines; these clusters are named a, b, c and d. In another variety of W. occipitalis occipitalis from the Austrian Alps (Figs. 4 and 5), the isolated spines Band D are lacking, spine A is rather small and spine C very long and slender; the specimens used for the figures were found on September 17, 1974 on the slopes of the Scheiblingstein, 850 m, not far from Lunzer Mittersee; they were on grass and bushes around a small spring. W. occipitalis occipitalis var. E seems to have a rather wide distribution in England and is not found in Continental Europe; in the phallus, spine D is lacking and spines E and F are small. Specimens of this variety were used by PICTET to describe Wormaldia occipitalis PICT., so that this variety is typical for the species. The W. occipitalis occipitalis described by L. BOTOSANEANU in 1960 and found by him in the Carpathian Mountains are very close to this typical variety from England, but spines A, B, D and E can each be split into 2 separate spines of the same size or of a different size and remaining parallel to one another. I described in 1974 four varieties of W. occipitalis occipitalis and a new subspecies of W. occipitalis, all from the French Alps. Undoubtably many more subspecies and varieties of this species will be discovered and maybe some more for other Wormaldia species. But one may call in question the value of such a classification taking into account minute characters; the only way to answer it is to find out whether there is or not genital segregation between the different varieties of a same subspecies. Can a male of one variety mate with a female of another, or vice versa, and are the eggs fertilized? For each population of Wormaldia occipitalis, emergence of the imagos occurs

26 I I / / / / I I / I I I / I

4

Figs. 1-3. Philopotamus ludificatus MeL., imago O. (1, 2) Specimen from the southwestern Alps (Mercantour mountain mass); (1) Phallus, ventral view; (2) Phallus, side view (on Figs. 1 and 2 the anterior part of the ventral plate has not been figured) ; (3) Specimen from eastern Austria, complex piece of the phallus, ventral view. Figs. 4 and 5. Wormaldia occipitalis occipitalis (PICfET), imago 0 from eastern Austria. (4) Phallus and right half of the genital parts; (5) Phallus, ventral side. 27 during several months of the year and, if one takes time and care, he can catch a few specimens at any time during that period; but there is a priviledged and short period of a few days, sometimes two or even three periods during which many imagos emerge simultaneously. Unfortunately priviledged periods of intensive emergence are not the same for the different varieties and this renders the tests of cross• breeding quite difficult to realize. I was able to make only one attempt of this kind between two varieties var. A and var. B of Wormaldia occipitalis occipitalis. The Isere valley, near the city of Grenoble, has on one side limestone mountains and, on the other, mountains made mainly of igneous rocks. In a certain spring (at Revel) on the igneous side develop, in large numbers, larvae of W. occipitalis occipitalis var. A; they live among dead leaves soaked in water or in wet moss. On the opposite side of the valley, in the limestone mountains, at about a distance of 9 miles from the other spring is one (at Claix) depositing calcium carbonate and flowing on a turf; the thick crust covering the rock is full of holes, channels and tiny galleries; the larvae of variety B build their silken nets within the galleries of the turf. The environment is thus very different for caddis-flies of one variety and for those of the other, during their immature stages; nevertheless the differences between the male imagos of variety A and those of variety B appear to be of little importance; the number of large isolated spines of the phallus is the same for both varieties and so is the number of clusters; but the number of spines for each cluster is quite different; besides the position of the forks on the wings is not at all the same. Fortunately the emergence of the imagos occurs about at the same time of the year on both sides of the valley and the priviledged period of intensive emergence for one variety overlapped - at least it did so once - the priviledged period for the other variety. I collected as many larvae ready to pupate of both varieties as I could, kept them in running water and obtained imagos of both sexes; I had plenty of specimens of variety A, but a much lesser number of caddis-flies of variety B; I prepared a large jar covered with a piece of cloth and containing branches and wet stones on its bottom; day after day I put in the jar males and females of variety A soon after they had emerged; the insects were active most of the time and mating between two individuals was seen quite often; later the females laid egg masses under the wet stones; I put these in running water and some time afterwards the eggs began to darken and young larvae could be seen inside. During that same period, I used two other jars, with branches, wet stones and a cover; in one I put males of variety A and a lesser number of virgin females of variety B; in the other I put males of variety B and about the same number of virgin females of variety A; in both these jars, when I was not handling them, the caddis-flies stayed quiet and moved only occasionally; nevertheless attempts to mate were seen and a few appeared to be successful; in both these two jars, almost all the females died without laying anything at all; two small masses of mucus without eggs were found; but one egg mass was layed and there were eggs in it; the stone it was layed on was put in running water" but not' a single larva did develop. Unfortunately this crossbreeding experiment was carried out with only a small

28 number of specimens of variety B, only 18 males and females altogether, so that no definitive conclusion can be given. It seems nevertheless that the two varieties A and B are already unable to crossbreed and that there is now a complete genital segregation. If we accept L. CUENOT'S definition of the species, variety A does not belong to the same species as variety B. If all the varieties of Wormaldia occipitalis of Europe - and there may be hundreds of them - are genitally segregated, as variety A and variety B are, they really are all distinct species; but this may not be true, at least for some of them. The experiment of crossbreeding I carried out is significant at ·least in one respect: The separation between the two varieties A and B of Wormaldia occipitalis occipitalis is a very recent event; it is possible that the male chromosomes of one variety can no more be accepted by the ovulae of the other, but nevertheless mating between a specimen of one variety and a specimen of the opposite sex of the other variety is still possible and this can even lead to the laying of an egg mass. Wormaldia occipitalis seemingly represents a very rare case, at least in Europe, among the caddis-flies. The larvae of this species develop mostly in springs that can be at a far distance from one another; geographical isolation has led to microendemism. But larvae of many other species of Trichoptera, belonging to the families Hydrop• tilidae, Rhyacophilidae, Psychomyiidae and Beraeidae, live also in springs or on dripping rocks; nevertheless these species seem to be perfectly distinct from one another and to have a large geographical area.

References

BOTOSANEANU, L. 1960. Revision de quelques especes de Philopotamus Leach et de Wormal• dia McL. (Trichoptera, Philopotamidae). Acta Soc. Ent. Cechoslov., 57: 223-228. CUENOT, L. 1936. L'espece, Paris. DECAMPS, H. 1967. Introduction 11 I'etude ecologique des Trichopteres des Pyrenees. AnnIs Limnol., 3: 101-176. VAILLANT, F. 1974. Quelques Trichopteres Philopotamidae de France et d' Algerie. Ann. Soc. ent. Fr. (N.S.), 10: 969-985.

Discussion

MORETTI: Vous avez fait des recherches sur I'habitat des especes de Wormaldia. Y-a t'il de gran des differences?

VAILLANT: ~ui, effectivement, il y en a qui vivent dans Ie genre d'habitat que M. Ie Dr. WIGGINS a decrit dans sa communication et qui est constitue par une bouillie brun noiratre, formee par des feuilles en decomposition et qui se trouve en bordure de certaines sources en sous-bois. D'autres se developpent parmi les feuilles mortes et la mousse en bordure de sources. Les autres sont franchement madicoles et, parmi elles, celles de certaines especes preferent les tufs de sources incrustantes. II semble y avoir un caractere commun 11 tous ces biotopes, la tres faible profondeur de I' eau. 29 MORSE: In my research I have had considerable difficulty everting the phallic membranes in my study material. I noticed that you have some excellent examples of everted phallic membranes. How did you accomplish this? VAILLANT: Here is my method: Put the whole caddis-fly in caustic potash; allow it to boil for about 10 minutes, or more if the concentration is low; take the insect out and drop it in boiling distilled water. Osmotic pressure makes the phallus come out. Then put in diluted acetic acid, wash several times and mount, after deshydratation, in Canada balsam. HANSELL: I was not quite clear about your crossbreeding experiments. I understood you had the var. A males with the var. B females. Did you have also any comparison between var. A males and var. A females, and var. B males and var. B females? VAILLANT: I indeed should have put together males and females of variety B, as I did for variety A, to make sure viable eggs could be laid in the covered jars. But I did not have enough live specimens of variety B; I had to use all of them for cross-breeding experiments. HIGLER: I must say this is an exciting lecture you gave, and I think this is one of the examples in which ecologists are reliable of the taxonomists to find out what all these species differences indicate. I was thinking of a suggestion. Dr. BOTOSANEANU told me some years ago that this is a typical case for cytological research in order to try to find out differences in these varieties in a cytological way. VAILLANT: I agree with Dr. BOTOSANEANU. It is indeed possible to compare chromosomes and chromosome numbers between different varieties of Wormaldia and this may lead to a much better understanding of the situation. BOTOSANEANU: A first problem. In your opinion pullus and copiosus are species of Dolophilus. I was sure they are Wormaldia too. VAILLANT: I found no differences between the larvae of Dolophilus and those of Wormaldia. The adults of the first genus differ from those of the second especially by characters of the wing venation, but also by the presence of a dorsal projection on abdominal segment X. I agree that this may not be enough to consider Dolophilus and Wormaldia as distinct genera. BOTOSANEANU: Et maintenant il s'agit d'un probU:me europeen: la variabilite de Wormaldia occipitalis. A mon avis, il s'agit d'un cas exceptionnel de variabilite qui meriterait d'etre etudie it l'echelle europeenne par un collectif. Le probleme a ete aborde par KIMMINS, VIGANO, par vous-meme et par moi, mais parfois avec une optique differente, et les resultats ne concordent pas. Le probleme est passionnant aussi du point de vue de l'evolutionisme. Si on reussit it former un collectif qui se propose d'etudier la variabilite de W. occipitalis sur l'ensemble de son areal, je pense qu'on peut arriver it des conclusions passionnantes du point de vue de la theorie de la speciation. VAILLANT: Inside a same population of Wormaldia, a particular cluster is not always made of the same number of spines; the number varies between two close limits such as 5 and 7 or 7 and 9. BOTOSANEANU: A problem more. I could not understand whether in your opinion all the varieties of Wormaldia occipitalis should be considered as species? VAILLANT: Yes, I believe that, if specimens of two varieties cannot breed together, these varieties really represent distinct species, even if morphological differences are small. BOTOSANEANU: I suppose it will be better to start from the opinion that W. occipitalis is a very variable species.

30 MARLIER: Could you observe also differences in the females of the different populations? VAILLANT: No, I could not find any differences between females of different varieties; I am only able to separate those of Dolophilus from those of Wormaldia on account of the wing venation. MALICKY: I have a problem with Wormaldia triangulifera in Greece. There is a certain number of spines in the aedeagus, and the complete set is always found in a certain percentage of specimens in every population. But the percentage of the specimens with the full number of spines differs from island to island. It is not possible to identify one specimen alone as belonging to a defined population. But if one has about ten or twenty specimens it would be possible to say from where the sample comes. STATZNER: What is your opinion about the function of these spines? VAILLANT: When the phallus is stretched out at its utmost length, all the spines are probably sticking outside, as they are permanently doing for a Tinodes. I do not really know what they are used for. For an identification of the species or of the variety it is better if the spines are still inside; their respective position can be seen more easily. CRICHTON: I was interested to hear that you boiled the specimens in caustic potash to bring out these structures. May I make the comment that it is enough to soak in cold caustic potash for a few hours; there is no need to boil. VAILLANT: Yes, you can obtain better slides in that way,-but it takes more time.

31 Proc. of the First Int. Symp. on Trichoptera, 1974, Junk, The Hague

Athripsodes BILLBERG and Ceraclea STEPHENS, distinct genera of long-horned caddis-flies (Trichoptera, Leptoceridae) 1

JOHN C. MORSE 2 & IAN D. W ALLACE3

Abstract Based on independent studies of immature stages and adults of Athripsodes, sens. lat. (Lep• tocerus auctt. nec LEACH, 1815), two distinct genera are recognized, Athripsodes BILLBERG, 1820, and Ceraclea STEPHENS, 1829, which together comprise a new tribe. Characters are given to distinguish the two genera in the larval, pupal and adult stages. Data are presented which support monophyletic hypotheses for the tribe and for each of the two genera.

MOSELY (1939), in his key to the British species of caddis-flies, distinguished two main groups within his genus Leptocerus (nec LEACH, 1815) on the basis of the presence or lack of flexibility of the fourth segment of the adult maxillary palp. In her treatment of the immature stages of Russian species of the same genus (Athrip• sodes, sens. lat.), LEPNEVA (1966) separated two groups along the same lines. Critical studies of the larvae of all the British species of this complex by one of us (WALLACE) and of the adults of most of its world fauna by the other have provided additional evidence for the distinctiveness of the two groups. The present work enumerates the morphological and behavioural characters useful for recognizing members of the two groups, referred to here as the genera Athripsodes BILLBERG, 1820, and Ceraclea STEPHENS, 1829, and notes the derived character states which indicate the probability of the monophyly of each. We are especially grateful to Dr. HERBERT H. Ross of the University of Georgia, U.S.A., and to Dr. G. N. PHILIPSON of the University of Newcastle-upon-Tyne, U.K., for their counsel and concern during our respective researches.

1 This study was supported by a grant for improving doctoral dissertation research in systema• tics from the National Science Foundation, by a grant-in-aid of research from Sigma Xi, The Scientific Research Society of North America, and by a special research grant from the National Environment Research Council of Great Britain. 2 Department of Entomology and Economic Zoology, Clemson University, Clemson, South Carolina, U.S.A. 3 Merseyside County Museums, William Brown Street, Liverpool 1, England.

33 Athripsodini, new tribe

The two genera Athripsodes and Ceraclea comprise a new tribe, Athripsodini, of the subfamily Leptocerinae. The adults of species in this tribe have tibial spurs arranged 1, 2, 2 or 2, 2, 2 on the three legs, respectively. The branching of the median vein in the forewing is typically stalked and the females of most species have an additional branch of this vein not present in the males. The stems of both the sectoral vein and the median vein are present in the hindwings. Cubitus vein is forked in the usual manner in the hindwing and the anal region of this wing is generally broader than in most other members of this subfamily. . The known larvae and pupae of members of this tribe have gills with several filaments. The gills are single in all known immature stages of other species in the subfamily. The monophyletic nature of this tribe is supported by the unique presence of a second pair of phallic parameres in the adult male (Fig. 20) and by the peculiar development of a pair of mesonotal bars in the larvae (Figs. 6 and 26). Nearly one quarter of the described species of leptocerids belong to this tribe. Lists of included subgenera and species will appear in the completed revisions.

Athripsodes BILLBERG, 1820 Type species: Phryganea albifrons L., 1758, subsequent selection by MILNE, 1934.

Athripsodes BILLBERG, 1820: 94. MILNE, 1934: 18. KIMMINS, 1949: 201. Mystacide PICT., 1834: 162, partim. Several succeeding workers in this or emended usages. Leptocerus WALKER, 1852: 57, partim. Many succeeding workers. The generic name Athripsodes (Greek: 'not like a thrips'. Gender masculine) was uncovered in BILLBERG'S rare book by MILNE after 114 years of omission from systematic literature. At the time of MILNE'S publication, all caddis-fly systematists were referring species of the group to Leptocerus LEACH. The monobasic type species of Leptocerus, Phryganea interrupta F., was generally included in Setodes RAMBUR (KIMMINS, 1949). Most workers now agree that Athripsodes (here Athrip• sodini), Mystacides BERTHOLD, Leptocerus and Setodes are distinct groups within the Leptocerinae. Distinctive character states of Athripsodes adults include the externally obvious midcranial sulcus (Fig. 1, mcs), the completely sclerotized, inflexible, fourth maxillary palp segment (Fig. 2), and the deeply divided tenth tergum of the male (Fig. 3). The larvae of Athripsodes species are markedly different from those of Ceraclea. The larval head of Athripsodes is always longer than broad and parafrontal lines are always absent (Fig. 4). The submental apotome ('gular sclerite') is always triangular (Fig. 5). The mesonotal bar in the final instar is short and straight (Fig. 6). The reinforcing sclerites (lateral case-holding plates) of the first abdominal segment are

34 each strongly bent in the distal anterior part (Fig. 7) or are slightly forked or with a large clear space in the anterior half (Figs. 8a,b). Gills are present on no more_than abdominal segments 1-3. No posterio-lateral projections are evident on the ninth abdominal tergum and a definite tergite is present (Fig. 9). There is a single, strong, accessory claw above each main anal claw (Fig. 10). The larval case is long, tapering, and slightly curved (Figs. 11 and 12). Its anterior dorsal lip never overlaps the ventral one, even in the earlier instars. Second instar larvae lack a row of long setae on the hind legs (Fig. 13). The pupae of Athripsodes species each have a triangular-shaped process or distinct bulge on the anterior median edge of the labrum (Fig. 14). Gills are arranged as in the larvae. The apex of each of the pupal anal rods is complicated and ends abruptly without tapering (Fig. 15a,b). The pupal case has round or narrowed vertical anterior and posterior openings (Figs. 16a,b and 17a,b). In the adult male of the hypothetical ancestor of Athripsodes, the subapico-dorsal process of the basal clasper segment became reduced in size with an increase of sclerotization (Fig. 18). Athripsodes aterrimus (STEPHENS) (Fig. 19) exhibits an intermediate stage in this modification from the ancestral character state (Fig. 39). The last instar larva and pupa of this ancestor lost the gills on abdominal segments four to eight. Ceraclea STEPHENS, 1829 Type species: Phryganea nervosa FOURCROY, 1785, monobasic (a synonym of Ceraclea nigronervosa (RETZIUS), 1783). Ceraclea STEPHENS, 1829: 28.

The generic name Ceraclea (derivation or meaning unknown. Gender feminine according to usage by STEPHENS.) was first published by LEACH (1815) without description or included species. The name was part of STEPHENS' list of British insects and, in his compilation, included only the species 'nervosa, Lat'. (P. nervosa FOURCROY). The name has not been given to designate a separate genus since 1885 and has never been applied to more than the one species before now. We recognize it here as the oldest available name for the genus-group whose characteristics are discussed below. The adults of Ceraclea species are distinguishable by the absence of a midcranial sulcus (Fig. 21), by the flexibility of the apex of the fourth maxillary palp segment (Fig. 22) due to a mottled loss of sclerotization, and by the odd number of projections of the tenth tergum, indicating the fused condition of the main tergite (Fig. 23). The larvae of members of this genus differ from those of Athripsodes species by having a head which is rather broad in relation to its length, particularly in the last instar, and by the usual presence of parafrontal regions (Fig. 24). The submental apotome of the last instar larva is short and broad, forming a trapezoid or barrel• shaped sclerite (Fig. 25). The mesonotal bars of final instar larvae are long and bent 35 Figs. 1-8,20-27. Character differences between Athripsodes (1-8) and Ceraclea (20-27) adults and larvae. (1, 4, 8a,b) A. aterrimus STEPHENS; (2) A. cinereus CURT.; (3, 5, 6, 7) A. albifrons (L); (20) c. minima KIMMINS; (21) C. transversa HAGEN; (22) c. tarsipunctata VORHIES; (23) C. nepha Ross; (24-26) C. annulicomis STEPHENS; (27) C. fulva RAMBUR; (1, 21) adult head, dorsal view; (2,22) maxillary palp of adult (shown without setae); (3,23) ninth and tenth terga of adult male, dorsal view; (4, 24) head of final instar larva, dorsal view; (5,25) head of final instar larva, ventral view; (6, 26) right half of larval mesonotum, dorsal view; (7, 8a,b, 27) left reinforcing sclerite on first abdominal segment of larva, lateral view; (a) forked form, (b) form with anterior clear space; (20) phallus, left lateral view.

36 369

Figs. 9-15, 28-36. Character differences between Athripsodes (9-15) and Ceraclea (28-36) larvae and pupae; (9, 14) A. aterrimus; (LO, 15) A. cinereus; (11) A. albifrons group; (12, 13) A. albifrons; (28) C. albimacula RAMBUR; (29) c. senilis BURMEISTER. (30, 35) C. fulva; (31, 33, 34, 36) C. dissimilis STEPHENS; (32) c. annulicomis; (9, 28) ninth abdominal tergum of larva, dorsal view; (10, 29, 30) larval anal.claw, lateral view; (11, 31) case of second instar larva, lateral view; (12, 32) case of final instar larva, la.teral view; (13, 33) right hind leg of second instar larva, posterior view; (14, 34) pupal labrum, dorsal view; (15a,b, 35a, 36b) pupal anal rods; (a) dorsal view, (b) lateral view of left rod.

37 about midway (Fig. 26). The reinforcing sclerites on the first abdominal segment are either straight or slightly curved in the anterior distal part (Fig. 27). Gills are present on at least abdominal segments 2-6, and sometimes 1 and 7 and 8. All instars have a pair of gill-like, filamentous, posterio-lateral projections on the ninth tergum and no distinct tergite is present, with only a weak, pigmented patch evident in some species (Fig. 28). Either one (Fig. 29) or two (Fig. 30) small accessory claws lie above each main anal claw. The larval case of Ceraclea species is typically of a cornucopia shape, especially in the later instars (Fig. 32), with a definite overhanging anterio-dorsal lip often recognizable even on cases of second instar larvae (Fig. 31). The hind legs of second instar larvae are each equipped with a row of long setae (Fig. 33). The larva uses its fringed hind legs to swim or at least to lift itself off the substrate. These setae are haphazardly shed late in that instar. The pupae of Ceraclea species do not have a triangular-shaped process or bulge on the leading edge of the labrum (Fig. 34). Abdominal gills are generally arranged as in the larvae. The apex of each of the pupal anal rods is tapered, usually from about midway (Figs. 35a and 36b). The pupal case has closing membranes with horizontal openings (Figs. 37a and 38a). Larvae of several species of this genus feed on whole particles of freshwater sponge and incorporate such particles and spicules into their cases. In fact, some

Figs. 16-19, 37-39. Character differences between Athripsodes (16-19) and Ceraclea (37-39) pupae and adults; (16a,b and 19) A. aterrimus; (17a,b) A. bilineatus (L.);. (18) A. albifrons; (37a) C. dissimilis; (38a) C. fulva; (39) C. cancel/ata BETIEN; (16a,b, 17a,b, 37a) and (38a) closing membranes of pupal case; (a) anterior membrane; (b) posterior membrane; (18, 19, 39) left clasper of adult male, lateral view.

38 species, C. fulva and close relatives, seem to require sponge in their diet in order to complete their life cycle. Peculiar developments which support an hypothesis of monophyly for this genus include (1) the loss of the midcranial sulcus and (2) the mottled loss of sclerotization on the fourth maxillary palp segment of the adult, (3) the fusion of the two halves of the tenth tergum of the adult male, (4) the broadening and (5) the development of the parafrontal regions of the larval head, (6) the change of shape of the larval, submental apotome discussed above and (7) the numerous known examples of species in Ceraclea with the otherwise exceedingly rare capacity to ingest solid sponge. In conclusion, it seems highly probable that the new tribe Athripsodini and the two genera Athripsodes and Ceraclea are each monophyletic groups as evidenced by their respective possession of unique character states listed above. We are convinced that the two genera indeed should be recognized as distinct, named groups of species because of the large number of morphological and behavioral differences between them. These differences are at least as great as those used to separate other genera in the family and the order.

References

BILLBERG, G. J. 1820. Enumeratio insectorum in museo GUST. JOH. BILLBERG. Gadelianis, Stockholm. FOURCROY, A. F. 1785. Entomologia Parisien sis, sive catalogus Insectorum, quae in agro parisiensi reperiuntur 12. 2: 234-544. KIMMINS, D. E. 1949. Some changes in generic names in the family Leptoceridae (order Trichoptera). Entomol. 82: 201-204. LEACH, W. E. 1815. Entomology. In D. BREWSTER (Edit.): Edinburgh Encyclopedia. 9(1): 57-172. LEPNEVA, S. G. 1966. Larvae and pupae of Integripalpia, Trichoptera. Fauna of the U.S.S.R., Vol. 2, No.2 (Trans., Israel Program Sci. Trans., Inc., 1971). Zool. Inst. Akad. Nauk S.S.S.R., New Ser. 95: 1-560. LINNAEUS, C. 1758. Systema naturae per regna tria naturae secundum classes, ordines, genera, species cum characteribus, differentiis, synonymis, locis. Tomus I regnum animale. Edito decima. Sumptibus Guilielmi Engelmann, Lipsiae. MILNE, L. J. 1934. Studies in North American Trichoptera. Privately printed, Cambridge, Massachusetts. 1: 1-19. MOSELY, M. E. 1939. The British caddis flies (Trichoptera); A collector's handbook. George Routledge & Sons, Ltd., London. PICTET, F.-J. 1834. Recherches pour servir a l'historie et a l'anatomie des Phryganides. Geneva. RETZIUS, A. I. 1783. Caroli Lib. BAR. DE GEER. Apud Siegfried Lebrecht Crusium, Lipsiae. STEPHENS, J. F. 1829. The nomenclature of British insects. Baldwin and Cradock, London. WALKER, F. 1852. Catalogue of neuropterous insects in the collection of the British Museum. London. 1: 1-192.

39 Discussion NIELSEN: It is not a question, but I have actually seen at lakeshores small swimming larvae, probably belonging to the genus Ceraclea. MORETII: Cette distinction des genres me semble tres bien fondee. Les caracteres que vous avez demontres sont comprehensants, n'est-ce pas? MARLIER: What about the 'genal suture' of some larvae of Athripsodes? Is that a primitive character? MORSE: No, it is not. The parafrontal lines developed in the larva of the ancestor of Ceraclea and nowhere else in the Leptoceridae. The frontal sulci are certainly primitive features and the pair of sulci which lie before and below the eyes are common throughout the Leptoceridae, but the parafrontal regions are a derived character state for the genus Ceraclea. CRICHTON: I wonder if you have any comments on the flight activity and swarming behaviour of these leptocerids? MORSE: I saw at one time a very exciting early evening flight with great numbers of Ceraclea, but I have not observed flights in a comparative fashion for the two genera Ceraclea and Athripsodes. RESH: The adults of Ceraclea appear to behave very similar to the European Athrip• sodes.Both are early evening fliers, and in Ceraclea, the evening flight activity of the males preceeds that of the females. MORSE: My completed revision of the genus Ceraclea is now ready for press. There are 87 species now known, and the evolutionary patterns for the subgenera Ceraclea and Athripsodina have been worked out. I thought it would be interesting to those in attendance here for me to mention which species are actually involved in the two genera. The genus Athripsodes does not occur in North America, all Nearctic species belonging to Ceraclea. The European species of Athripsodes are albifrons (L.), aterrimus (STEPHENS), bilineatus (L.), braueri (PICTET), cinereus (CURTIS), commutatus (ROSTOCK), cuneo rum (MACLACHLAN), fulvoguttatus (MOSELY), genei (RAMBUR), inaequalis (MACLACHLAN), interjectus (MACLACHLAN), leucophaeus (RAMBUR), longispinosus (MARTYNOV), rieli (NAVAS), and tavaresi (NAVAS). The European species of the nominate subgenus of Ceraclea are albimacula (RAMBUR), alboguttata (HAGEN), fulva (RAM• BUR), nigronervosa (RETZIUS), nygmatica (NAVAS), senilis (BURMEISTER), and a species yet unnamed. The European species of Ceraclea subgenus Athripsodina are annulicornis (STEPHENS), aurea (PICTET), dissimilis (STEPHENS), excisa (MORTON), norfolki (NAVAS), perplexa (MACLACHLAN), riparia (ALBARDA), and sobradieli (NAVAS).

40 Proc. of the First Int. Symp. on Trichoptera, 1974. Junk, The Hague

Communication sur trois larves de Trichopteres du Nepal (Progress report)

LAZARE BOTOSANEANU

II s'agit de trois jolies larves de Himalaya du Nepal, que je viens de decrire. Ce que vous voyez c'est Ie fourreau d'un Glossosomatide, il s'agit d'une construction tres delicate de 3-4 mm de longueur, tres fine, il y a une ressemblance generale avec les fourreaux de Synagapetus, mais chez Synagapetus nous avons un chapeau avec un bord continu, tandis qu'ici Ie bord du chapeau est remplace par quatre anneaux. C'est une construction tres remarquable a mon avis. L'etude morphologique de la larve montre que c'est un Catagapetus, mais Ie fourreau est tout a fait different de celui de Catagapetus nigrans d'Italie, avec ses deux 'cheminees'. lei il n'y a aucune trace de cheminee, mais bien deux fentes longitudinales sur la bosse ellipsoid ale qui couronne Ie fourreau. lei vous voyez les fourreaux-piege d'un Diplectrona. C'est une construction de 4 ou 5 cm, tres solide; vous voyez qu'il s'agit d'abord d'une sorte de tunnel qui est applique sur les pierres des torrents. L'aspect ventral du fourreau detache de la pierre a l'aide d'une lame, permet de voir une 'levre' de 5 ou 6 mm de large qui assure l'adhesion du fourreau sur les pierres. Pres de l'extremite posterieure sur la face dorsale il y a toujours un orifice. En angle droit avec Ie fourreau proprement dit vous voyez une sorte de 'cou' dont l'orifice est represente ici par transparence du filet; ce 'cou' est en relation avec un cadre compose de materiaux vegetaux melanges de secretion; ce cadre est tres soli de, mais il est flexible. Les deux comes formant Ie cadre peuvent s'eloigner et se rapprocher l'une de l'autre, ce qui represente une interessante adaptation a la vie dans les torrents impetueux de I'Himalaya, parce qu'avec un cadre rigide Ie filet risque d'etre detruit par Ie courant. Et puis, tendu sur Ie cadre, il y a un filet d'une construction extremement soignee, avec des mailles extremement regulieres et de la solidite d'un fil de nylon. Et ici quelque chose qui n'est pas tout a fait nouveau puisque les constructions des Uenoidae ont deja un peu ete decrites du Japon. C'est Ie premier Uenoidae de l'Himalaya. Vous voyez que les fourreaux, longs de 3 cm environ, sont entierement composes de secretion verdatre. lIs sont translucides et composes d'un nombre tres variable d'anneaux. II serait tres interessant d'etablir une correlation entre developpement postembryonnaire et nombre des anneaux: moi, je n'ai pas reussi. Les fourreaux nymphaux sont obtures a l'extremite anterieure par un opercule, et a

41 l'extremite posterieure par un autre (mais pas tout a fait a l'extremite posterieure, parce que la position de cet opercule est anteapicale). Les fourreaux sont groupes en grand nombre, ils sont attaches a l'aide d'une amarre tres solide de secretion noire a des plaques de secretion qui recouvrent les pierres du torrent.

Discussion

FLINT: I would like to make an observation on the last Himalayan larva that you showed. I just recently received several collections from Nepal, that I believe are the same thing. It is very close to Neothremma of North America. The cases of these two are very different but the larval morphology is very similar. BOTOSANEANU: You know, the Uenoinae are now considered as subfamily of the Threm• matidae. In my opinion this is wrong. I can see no relation between Uenoa and Thremma. The European workers know Thremma, it is completely different. The thorax of Uenoa is com• pletely different from Thremma. About Neothremma I know almost nothing. FLINT: Yes, I am certain that that is the larva I have. BOTOSANEANU: You are also of the opinion that Uenoa is strongly related to Neothremma? WIGGINS: Yes. FLINT: I think this is another genus of the same group. I don't know if it is the same genus though. BOTOSANEANU: It is hard to understand that some members of the same family build such cases, completely secreted, transparent, and made of rings, whereas some other construct cases of sand, very typical. FLINT: Even in one genus you can find that, e.g. in Brachycentrus in North America. Sometimes, within the same species one will find a circular case made completely of secretion, and other cases, the typical square ones of plant material. BOTOSANEANU: I agree with you, that only the morphology will show the exact relationships. On the sides of the 8th abdominal segment of the larva I have found a number of 18-25 flexible appendages having somewhat the appearance of trachea. They seem to be very characteristic. I don't know what they represent, having a complicated spiral structure. In Thremma they don't exist. But this is not so important. The most important is the structure of the thorax.

Ross: In one species of North American Brachycentrus one fre~uently finds a case in which the bottom part is square, then there is a round secretion part, and about that another square portion, two types in the same individual. Usually the case is all square, but in some populations you get perhaps 25% having this type of mixed construction: WIGGINS: There are really two genera in North America that follow this same characteristic, Neothremma and Farula. It is not possible in the literature to distinguish Farula from Neothremma yet, but Farula larvae frequently make cases entirely of secretion. It's not unusual. FLINT: I think MOSELY described this Asian genus in one of his works on Indian caddis-flies. I have only one poor female in my material, and on a quick look through MOSELY'S works, I found a figure that looked like mine. I just can not remember the name now.

42 WIGGINS: Eothremma? BOTOSANEANU: Exactly! But he says nothing about the larvae. FLINT: No, nothing on the larva. But it was another genus: Ashmira. BOURNAUD: Savez-vous ce que mange Ie Dipleetrona que vous avez montni? BOTOSANEANU: J'ai un materiel tres riche. Tous les filets sont bourres de larves de Chironomides; il y a d'autres Dipteres ayssi (Simuliides), quelques Copepodes. MALICKY: They had it also in the intestine? BOTOSANEANU: I don't know, I never made dissections. Probably. MARLIJ;R: QueUe est la dimension des mailles du filet? BOTOSANEANU: Je ne I'ai pas mesuree. Vous pouvez vous faire une idee: Ie diametre du filet des exemplaires les plus grands est de 1.5 em.

43 Proc. of the First Int. Symp. on Trichoptera, 1974. Junk, The Hague

A progress report on the endemic element of Tasmanian Trichoptera

ARTURS NEBOISS

A revision of Tasmanian Trichoptera, based on material collected during 1965- 1972 period is now in progress and preliminary information indicates that the number of species known from this region will increase from 85 to more than 145, or over 70%. For the analysis of distribution and endemic content of Tasmanian Trichoptera the island is subdivided into ten provinces. The subdivision is based on a combination of landforms, vegetation and climate. A preliminary count of species known to be endemic indicates that there are about 100 or 68% in this category. This figure is quite close to that found in some other insect orders. As yet it is impossible to finalize any figures because the Australian mainland fauna, particularly that of Victoria, has not been sufficiently investigated. From the data on distribution it became obvious that there are certain areas with large numbers of endemic species, whereas in others the proportion appeared somewhat smaller. Not all provinces have been sampled equally thoroughly, but even so, there is an obvious, quite definite trend. The analysis of recently collected material, to which the information from earlier publications has been added, clearly shows that the Central plateau and the South-western province have the highest proportion, about 75%, of endemic species, closely followed by the North-western province and Southern• forests with over 60% decreasing in the eastern provinces to 50% or even less.

Discussion

ILLIES: This distribution of percentages of endemics is similar in the Tasmanian Plecoptera fauna.

45 Proc. of the First Int. Symp. on Trichoptera, 1974. Junk, The Hague

A preliminary report of studies on neotropical Trichoptera

OLIVER S. FLINT, JR.

Studies commenced in 1961 with field work in Puerto Rico and in following years on Jamaica, Hispaniola, Dominica, St. Lucia and Grenada. Starting in'1965 for three summers, field work was concentrated in Mexico and Central America. In recent years, I have visited Chile and Argentina five times, culminating in five and a half months in Argentina in 1973-1974. In addition, I have received large collections for study from Surinam, The Amazon Basin and Uruguay. Preliminary studies are elucidating the zoogeographical boundaries involved. In Mexico, the Sierra Madre Occidental and Sierra Madre Oriental provide routes for the intermingling of elements of the Nearctic and Neotropical faunas. The Nearctic Fauna is limited southwardly by the Trans-Mexican Volcanic Belt, although a few elements occur on the high peaks as far south as Panama. The Neotropical faunal elements are only sparsely represented in the first tier of southwestern states in the United States. The Neotropical Region itself is sharply divided into two subregions, the Chilean and Brazilian. The Chilean fauna contains 15 endemic genera of Rhyacophilidae, and shares one genus with the Brazilian which contains one very large endemic genus. In the Sericostomatids there are nine endemic genera in the Chilean Subre• gion and one endemic in the Brazilian. In addition, the families Rhyncorheithridae, Tasimiidae and Kokiriidae are limited to the Chilean area in the New World. All these elements are similar to those of Australia and New Zealand. Within the Chilean Subregion there do not seem to be any clearly defined subpatterns of distribution. In the Brazilian Subregion, however, several common distributional patterns are found. One encompasses most of Mexico and the drier areas of Central America into northwestern Costa Rica. The Greater Antilles possess a rather distinctive fauna which primarily seems to be derived from Mexico. There are at least two genera, however, the polycentropid Antillopsyche and the Zygopte• ran Phyllolestes which do not have any relatives in the New World but are clearly related to certain African forms. In the wetter areas of Mexico, Central America, and the Lesser Antilles one finds a more typical Brazilian fauna, often specifically the same as that of the Amazonian area. This distributional pattern is often referred to as the Circum-Caribbean Pattern. 47 There is one rather interesting area of the mountains in eastern Brazil that contains primarily a typical Brazilian fauna. However, it contains in addition, several endemic elements, e.g. Barypenthus, Grumicha and a Sortosa species, that are not closely related to anything else in Latin America. It is possible that these might be relicts of Gondwanan connection with South Africa.

Discussion

NEBOISS: Do you have any information on the South African, South American connections in the distributional pattern? FLINT: I have not investigated that at all. The South African fauna does not seem to be very closely related to that of southern South America. However, I do find that there are a few things recorded from South Africa that should be investigated from this aspect. ILLIES: What about the Chilean-Australian-New Zealand relationship? FLINT: Yes, this similarity is very marked in the Trichoptera, as in the other aquatic insects. In fact, the fauna of the Chilean Subregion is in general more similar to that of Australia-New Zealand than to that of the Brazilian Subregion.

48 Proc. of the First Int. Symp. on Trichoptera, 1974. Junk, The Hague

The distribution of the Hydropsychidae in 'Great Britain

RUTH M. BADCOCK

Larvae of the Hydropsychidae are generally known as net spinners, and they live in flowing water. They build fixed dwellings with a net incorporated in the vestibule, and a current of water is necessary for the spinning of the regular mesh of the net. Many of the species now occurring in Britain tend to be associated with upland, stony-bottomed streams and rivers rather than silted lowland courses with less oxygen and slower currents. A firm substratum is required for anchorage of the dwellings and for the eggs, which are cemented to submerged surfaces by the female imago who enters the water to oviposit (BADCOCK, 1953). The geographical distribution of the eleven species of the Hydropsychidae that are known to occur or to have occurred in Great Britain has been mapped. Thanks are due to Miss C. ALLEN of the Biological Records Centre of Monkswood Research Station for preparing the maps from the variety of data supplied to her. These comprised: a. My own records, mainly collection of larvae over twenty-five years - with frequent rearing to imagines where necessary - and localities for larvae sent to me for identification (Map representation - a black dot). b. Rothamsted Insect Survey Light Trap captures. These imagines were identified by Dr. M. I. CRICHTON of Reading University. I am grateful to him and to Dr. L. R. TAYLOR of Rothamsted for allowing me to use these data (Map representation - a black square). c. Museum specimens (imagines), the majority of which are in the British Museum (Natural History). Captures before 1940 (represented by a ring) and from 1940 onwards (represented by a black triangle) are differentiated. d. Records from literature, 1940-1971 (represented by a + sign). Records have been corrected in accordance with the recent changes in nomencla• ture. I follow DOHLER (1963), BOTOSANEANU & MARINKOVIC-GOSPODNETIC (1966), and TOBIAS (1972) in recognizing as Hydropsyche siltalai DOHLER the species which ULMER (1909) and MOSELY (1939) had regarded as H. instabilis, and in retaining the name Hydropsyche instabilis (Curt.) for the

MounU>n1 ana PI."",.. _ 100011 oaom )

_ "'9" - •. 700-1OOOft\:c> 1IO·0a0m) "'9"" 7OO-X)OOf' (,» 210 ' .00",) S ow Plot ...... !50'7,)(Itt (.. lOS·, '" ~ ./ i' o "' 350 .. JOOIt. OS'2I Om) 'f

:=:::J L ~ "" '"

Fig. 1. Relief map of England, Scotland and Wales. Reproduced, by kind permission of Prof. J. W. WATSON and Thomas Nelson & Sons Ltd., from: The British Isles, a Systematic Geography, edited by J. WREFORD WATSON with J. B. SISSONS.

50 distribution of some hydropsychid species falls within that of the more hilly areas. This can be seen on looking at a relief map (Fig. 1) and then at the distribution of Diplectrona felix MeL. (Fig. 2), a head stream species, or of Hydropsyche siltalai (Fig. 3) which is the commonest hydropsy chid species in trout becks of the cool upland regions of Scotland, the Pennines, the English Lake District, Wales and the south• west. This species is absent in lowland areas where rivers are slow flowing and muddy e.g. south-west Lancashire, the Yorkshire Wolds and most of E. Anglia (except where upland stretches of stony stream· of trout beck type occur). EDINGTON (1968) mapped the rate of flow and distribution of nets of Hydropsyche siltalai in a stream. He found very few nets where the current was less than 15 cm/sec and at the other extreme, the upper limit of flow where nets were found exceeded 100 cm/sec. When he reduced the current flow by placing a baffle across the stream, larvae left the areas of quiet waters thus created and moved to positions where the water flowed more rapidly, showing that the larvae selected situations with a rapid flow. It is not intended to publish the full series of maps here; that is expected in a British journal. Only some representatives will be depicted in this paper but points relating to the distribution of other species will be briefly mentioned. The geographical distribution in Great Britain of H,ydropsyche pellucidula CURT. is fairly similar to that of H. siltalai. These two species may occur in mixed populations but there tends to be a difference in the timing of peak populations of well grown (5th instar) larvae. In mid Wales and the Lake District, H. pellucidula emerges earlier than siltalai and one finds the next generation of 5th instar larvae present in autumn when siltalai is represented by early instars. The peak population for 5th instar larvae of siltalai is in mid-summer (June and July). PHILIPSON (1957) main• tains that pellucidula larvae dwell in slower flowing and deeper water than those of siltalai. In my experience they may often dwell in the same habitat, though at different times, but pellucidula does extend further down stream into larger rivers than does siltalai. The British distribution of Hydropsyche angustipennis CURT. (Fig. 4) is in marked contrast to that of H. siltalai and H. pellucidula, for angustipennis is noticeably rare in the uplands of the north and west, and is the prevalent species of the midlands and south (except at higher altitudes). Where it does occur in Scotland and northern England, it is in streams which are warmer in summer than are neighbouring ones e.g. because they are surface run off from sheltered, unshaded tams receiving considerable insolation. H. angustipennis larvae appear to be more tolerant of mild organic pollution and oxygen depletion than are other British hydropsychids. In my experience, it is the hydropsychid which travels best! In my early days, before I used muslin and moss packets in thermos jars to transport larvae, angustipennis would survive train journeys when larvae of siltalai died. AMBUHL (1959) showed that larvae of H. angustipennis could vary their oxygen consumption according to conditions. Altogether it is a fairly tolerant species, occurring in both streams and rivers, including situations with gentle flow, provided that the summer water temper• ature is adequate. 51 2-- - -*---~4---~---~6~---J,O

BloIogcal Records Centre

Fig. 2. Dipleetrona felix McL., recorded distribution in England, Scotland and Wales. Key to symbols used in Figs. 2, 3 and 4 • Larvae identified by R. M. BADCOCK • Rothamsted Insect Survey Trap captures (imagines), determined by Dr. M. I. CRICHTON. o Museum specimens (imagines) captured before 1940 ... Museum specimens (imagines) captured from 1940 onwards + Records from Literature 1940-1971. 52 0 1 2 3 4

~·'r u J/ + B~ HYOIO'SYCHE t5 . ~ ~ 9 ;. SILTAlA. D~·hl., c.t' <\(t .~ D HI . -1 ~ B . 8 ... "-4 ~- ., >CO, ~+ + ....~ -r¥ ...... , + 0 t '1" ~$t It-+ ": I.. • :) ~ 1 7 ,tif:~~ ~ 12 ",,- ,[jc ~. :'\ 8 E I!. i~ V • • ~O 0 ~ ~ ~;o. 0 • I ,~ 5 !p, _Ik,.7, rJ Iii ). I ~ . .~ ~ \ . 4 Q < ~ 0 •.. • .A A. i • . !~ ( M< . • &-~ d ~ • . 3 - ~ ... 1 • V V. • ~ • ~ /,£ do. r- 0" _. V bi 2 ~~ 0 ~ .... ~ O~ ~: ·0 Ir 9 0 ~ ~ . .. 'e! ./r~ • .'" 1 ...- . -o;;.,J _1'\.0"(00'11 •• ~ ,)IWc.IIIO <'.1-, ~ ~ ~ 0 1/ 0 1 2 3 4 5 8

BIOlogical Records Centre

Fig. 3. Hydropsyche sittalai DeIHLER, recorded distribution in England, Scotland and Wales.

53 0 I 2 3 4 ~"tl f:=j~ .n ~ T HYD.O'SV(HI ~ ,,, g AHGUSTlPfNNIS ICloul.) -l.. /, ' -1 "~t • -D HI f. 8 - ~ 8 .. ~6 ~• .! ...! k- ~ I~ ( ! ... ~q ';' l/iit I .... >/ 1 \ 'tv- r~ 11" ~rfJ.:! ...... ' K'· « ,{~'k: ,So )...... \ n 6 ( ,;., V ! ~ ' ! (\ ~r 0 ,/"' ) \", "t- .. ! b ff 5 . . ~. iJ , 111 6 • 0 + .~ 4 ~ ...!" f -.. ~> \, .~ i. , • .: ) " . ~ ·I .. /-J ~. / r •• t·· • 3 . ",",' I" . ~ , •• • ~ , "; , !o ~ . "" ---. ~: -;7 Ie .~ e •• -.~ ~ ) ~ ~i :~ J' ~'" I ..l •• -' ". • ° 2 Ib. . \, . " -' e' ( -. J o. ~ po "'do '9 0 '. 00 5 j 6 ./ . I -,""0"(001 • " I· ""I 101 QIIIC - <3 ...... , r..:. ( .. 0 0 I 2 3 4 S 6

BiOIogcal Fleco(ds Centre

Fig. 4. Hydropsyche angustipennis CURT., recorded distribution in England, Scotland and Wales.

54 In contrast, larvae of Diplectrona felix dwell near the source of cool, shaded, stony trickles and, as already indicated, the distribution of this species is associated with upland regions. Hydropsyche instabilis has a rather similar, scattered distribution but ecologically it succeeds Diplectrona down stream, although mixed larval populations may occur. The habitat of Hydropsyche instabilis ranges from small stony streams to those of trout beck size, where it may overlap with, then be replaced by, H. siltalai. EDINGTON & HILDREW (1973) showed that the oxygen consumption of H. instabilis was relatively independent of temperature over a range of 5-15 °C but that of D. felix only over one of 5-10 °C. Growth rates and respiration rates of the two species were compared in two sets of fluctuating temperatures, one of which frequently exceeded 15°C (summer warm conditions) and one which rarely did (summer cool). The respiration rate of H. instabilis was similar in the two regimes but growth was less in the summer cool water; food was being used less efficiently. Diplectrona felix was less efficient in the summer warm water, but its growth remained constant while the oxygen consumption rose. In Great Britain, Hydropsyche fulvipes CURT. is relatively rare and localised. In the two areas where I have taken it, the larvae were in small streams of head stream type. In one, H. instabilis occurred in the same stream but fulvipes emerged before instabilis. I was mistaken in stating (BADcocK, 1955) that fulvipes was widespread in Britain. This was due to confusion of fulvipes and instabilis. Unfortunately I did not see the papers by BOTOSANEANU & MARINKOVIC-GOSPODNETIC (1966) and by TOBIAS (1972) until 1973, so until then was using the old nomenclature and had regarded fulvipes and instabilis as one highly variable species which I called fulvipes. I had identified instabilis as fulvipes for EDINGTON & HILDREW, so their 1973 paper contains this error. I have corrected the nomenclature in my references in this paper. , Hydropsyche saxonica McL. is very rare in Great Britain. The first published record for it was that of GRENSTED (1943) for Headington, near Oxford and this still seems to be the only known locality for it. I reared larvae from the Bayswater Brook, near Headington, to saxonica in 1954 but have not seen saxonica larvae there since 1955. They disappeared due to the development of a housing estate, pollution and the dredging of the brook. The British Museum (Natural History) has GRENSTED'S imagines as its representatives of British H. saxonica, and also a male from the Harwood Collection but unfortunately the locality of this is only given as 'England'. The female imagines from Unst (Scotland) and Co. Mayo (Eire) proved not to be saxonica, although included in that section. Hydropsyche contubemalis McL. has a sporadic distribution, as does Cheumato• psyche lepida PICT. Larvae of both are in fact associated with large rivers (over 40 m wide), usually with stony beds. Larvae of C. lepida always seem to be very much less numerous than those of H. contubemalis and to require a diligent search but they may occur mingled with H. contubemalis. Sparse populations may account in part for the absence of Cheumatopsyche lepida from the Rothamsted Insect Survey traps even when the trap was near a large river and catching H. contubemalis. The reason 55 for the association of these two species with large rivers is still uncertain and needs further investigation. It may perhaps be linked with temperature fluctuations being less in a large volume of water (LANGFORD, 1970). There are no recent authentic British records of either H. exocellata DUFOUR or H. guttata PIer. H. exocellata has not been found in Britain since 1901 and the most recent authentic record of H. guttata was from Southampton in 1915, an imago now in the British Museum. Last century imagines of both these species were taken around the Thames in the London area but they may now have become extinct in Great Britain. To conclude, insularity presumably accounts for the absence of certain continental species from Britain and for the restricted number of species on small islands around the British coast, e.g. only H. siltalai and Diplectrona felix are recorded from the Isle of Man. The distribution of some species can be correlated with physical relief, and a partial geographical pattern may be detected in the distribution of H. angustipennis which preponderates in the midlands and south. Otherwise, it is probably true to say that local ecological factors are of more importance than are broad geographical factors in influencing the distribution of the Hydropsychidae within Great Britain. The biotic ecological factors of food and predation have not been discussed here, partly due to lack of time but also because it is considered that they are of more importance in determining the level of population density than presence or absence, e.g. particularly dense populations of Hydropsyche larvae often tend to be associated with lake out-flows or with hard waters, where food is more abundant. Where Rhyacophila dorsalis larvae (predators of Hydropsyche) abound, Hydropsyche may be less numerous. A downstream sequence can be demonstrated, with Diplectrona felix characteristic of head streams near their source, followed by Hydropsyche instabilis, with H. fulvipes in a few localities, then H. siltalai (which is not in head waters) and H. pellucidula. Finally, H. contubernalis and Cheumatopsyche lepida are normally confined to large rivers (into which H. pellucidula may extend). Hydropsyche angustipennis is restricted to warmer waters and may range from small streams to large rivers where the temperature is suitable.

Summary

The geographical distribution of the hydropsy chid species occurring in Great Britain has been mapped. The maps for Diplectrona felix MeL., Hydropsyche siltalai DOHLER and H. angustipennis CURT. are depicted here; points relating to other species are men• tioned. Some species, such as Diplectrona felix and H. siltalai, are particularly associated with uplands. H. siltalai predominates in trout becks of hilly regions, especially in the north and west, where H. angustipennis is infrequent. H. angustipennis is the common species in the midlands and south, except at higher altitudes. 56 Otherwise, in the main, local ecological factors are considered to be of more importance than broadly acting geographical factors in influencing distribution within Britain; the sequential downstream distribution of British hydropsy chid species is described and oriefly discussed.

References

AMBUHL, G. 1959. Die BedeulUng der Stromung als okologischer Faktor. Schweiz Z. Hydro!. 21: 133-264. BADCOCK, R. M. 1953. Observation of oviposition under water of the aerial insect Hydropsyche angustipennis (CURT.) (Trichoptera). Hydrobiologia 5: 222-225. BADCOCK, R. M. 1955. Widespread distribution in Britain of our allegedly rare caddis, Hydropsyche fulvipes (CURT.) (Trich. Hydropsychidae). Entomologist's mono Mag. 91: 30-31. BOTOSANEANU, L. & MARINKOVIC-GOSPODNETIC, M. 1966. ContIibUiion a la connaissance des Hydropsyche du groupe fulvipes-instubilis. Etude des genitalia males: Annis Limno!. 2: 503-525. DOHLER, w. 1963. Liste der deutschen Trichopteren. NachrB!. bayer. Ent. 12: 17-22. EDINGTON, J. M. 1968. Habitat preferences in net-spinning caddis larvae with special reference to the influence of water velocity. J. Anim. Eco!. 37: 675-692. EDINGTON, J. M. & HILDREW, A. H. 1973. Experimental observations relating to the distribu• tion of net-spinning Trichoptera in streams. Verh. int. Ver. Limno!. 18: 1549-1558. GRENSTED, L. W. 1943. The occurrence of Hydropsyche saxonica McL. in Britain, with a new key to the British species of the Genus Hydropsyche Pict. (Trich. Hydropsychidae). En• tomologist's mono Mag. 79: 35-38. LANGFORD, T. E. 1970. The temperature of a British river upstream and downstream of a heated discharge from a power station. Hydrobiologia 35: 353-375. McLACHLAN, R. 1874-1880. A monographic revision and synopsis of the Trichoptera of the European Fauna. London. MOSELY, M. E. 1939. The British Caddis-Flies (Trichoptera). London. PHILIPSON, G. N. 1957. Records of Caddis-Flies (Trichoptera) in Northumberland with notes on their seasonal distribution in J'lessey Woods. Trans. nat. Hist. Soc. Northumb. (New Series) XII: 77-92. TOBIAS, W. 1972. Zur Kenntnis europiiischer Hydropsychidae (Insecta: Trichoptera, II). Senckenberg. bio!. 53: 245-68. ULMER, G. 1909. Die Siisswasserfauna Deutschlands, Heft 5-6: Trichoptera. Jena.

Discussion

NIELSEN: Concerning Hydropsyche gutrata-ornarula-contubernalis: SVENSSON and TJEDER and NIELSEN worked on this problem, but could not say which of these species occurs in Denmark. Cheumatopsyche lepida has in Denmark been found only in the middle reaches of the river Gudenaa which here is about 50 m broad. It is still abundant there though the river is rather much polluted. FLINT: Are you able to separate the species of the larvae of Hydropsychids? BADCOCK: I think so now as regards the British ones. There has been confusion for a number of years, particularly relating to H. fltlvipes, H. instabilis and H. saxonica but I think I have sorted this out now. FLINT: Did you find any tendency to vary, in colour patterns or other characters you use? 57 BADCOCK: Yes, colour patterns are not entirely reliable. One has to use morphological features as well. ILLIES: In the Fulda we found 5 or 6 species of Hydropsyche, it was rather easy to distinguish them as adults and to fix the distribution, but it was a hard job to find out any good larval characters. Our old late friend SATILER studied 40 different characters to find it out and did not succeed. I an happy to hear that you found suitable characters. Is it possible to use the colour pattern, marks of the head, etc.? BADCOCK: Yes, they are a useful guide if confirmed by morphological features. Besides the colour pattern of the frontocylpeal apotome, one can sometimes use the colour of the mentum and submentum. CRICHTON: I wish to compliment Dr. BAD COCK on her paper. Later in this Symposium I shall be showing computer-produced maps from the Rothamsted Insect Survey which agree very well with her maps prepared at the Biological Records Centre at Monkswood. Ross: In North America we have a great deal of trouble identifying many Hydropsyche larvae. You might give us hints as to new characters we might examine. BADCOCK: One really needs good morphological features as well as colour patterns and must guard against their being environmentally induced variations. I had two larval populations, one with long and one with short tarsal claws, which I suspected of being different species. How• ever they proved to be the same species; the length of tarsal claw reflected the hardness of the substratum and the degree to which the claws were worn down!

58 Proc. of the First Int. Symp. on Trichoptera, 1974. Junk, The Hague

Les Trichopteres de I'espace carpato-balkanique, foumisseurs de documents pour I'etude de I'evolution

L. BOTOSANEANU

J'ai choisi parmi les documents fournis par les Trichopteres de l'espace carpato• balkanique, trois cas qui me semblent mieux approfondis et plus riches de significa• tion pour la Biologie Generale. Puisse ce travail stimuler les trichopteristes de tirer de ces insectes tout ce qu'ils peuvent donner a la theorie de l'Evolution, puisse-t-il montrer aux specialistes de la Biologie Generale l'interet considerable que presente Ie groupe, puisse-t-il enfin stimuler les specialistes de la genetique des populations d'approfondir a I'aide de leurs methodes des cas comme ceux que je vais presenter. Le premier cas est celui des especes du genre Annitella. Six especes de ce genre forment, par tous les details de leurs armatures genitaies 0 et <;2, une extraordinaire serie lineaire. Ces especes sont: A. lateroproducta BOTS., A. transsylvanica MUR• GOCI, A. dziedzielewiczi SCHMID, A. kosciuszkii KLAP., A. chomiacensis DZIEDZ., A. thuringica ULM. Dans cet ordre, toutes les pieces des genitalia forment des series progressives, sans exception aucune (Fig. 1); chez Ie 0 , Ie rapport entre la taille du 8e tergite et celie du sternite correspond ant se renverse graduellement en faveur du tergite; les angles latero-apicaux du 8e tergite sont initiellement en forme de lobes arrondis, non scIerotises, puis de simples pointes scIerotisees se forment, elIes se transforment en petites comes noires dirigees d'abord vers la ligne mediane et ensuite vers Ie haut et lateralement, et celles-ci revetent enfin l'aspect de tres longs appendices noirs fort recourbes; les appendices superieurs sont d'abord de minces bandes a peine elargies a I'apex, mais leur aspect capite devient de plus en plus evident; les "appendices intermediaires" se composent de deux branches: chez I'espece la plus primitive la branche mediane est nettement plus courte et plus mince que celie laterale, ce rapport se renverse graduellement et en fin de serie la branche mediane surpasse nettement celie laterale; les gonopodes diminuent de taille, leur partie apicale est petite et horizontale en debut de serie, puis s'allonge graduellement vers Ie haut; chez la <;2, la partie dorsale de I'armature presente une nette tendance a I'augmenta• tion de la taille; celle ventrale, au contraire, est de taille nettement decroissante dans la serie. Je vais maintenant plaquer sur la carte cette serie d'especes (Fig. 2): A. lateroproducta est connue de presque I'ensemble des Carpates de Roumanie; A. trans• sylvanica d'une seule localite des Mts. de Rodna, Maramouresch (Roumanie); A.

59 d

Fig. 1 Representations demi-schematiques des armatures genitales 0 vues de face (A, B) et des armatures genitales 'i' en vue laterale (C, D) chez les deux termes extremes de la lignee orthogenetique des Annitella: lateroproducta a gauche, thuringica a droite. 60 dziedzielewiczi des sources du Pmt, dans les Carpates de l'URSS; A. kosciuszkii des Mts. Czamohora et Chomiak, Carpates de l'URSS; A. chomiacensis, decouverte dans 1e bassin superieur du Pmt (Carpates de I'URSS), a ete retrouvee dans Les Beskides Orientales, en Pologne; A. thuringica possede un assez considerable areal: Carpates de Slovaquie, de Poiogne, Mts. de Moravie et de Boheme du Nord, Autriche Inferieure, Thuringe, Harz, Westphalie. Une conclusion s'impose aussit6t: les especes formant ceUe serie parfaitement rectiligne du point de vue morphologique, se succedent Ie long d'un axe pricipal qui epouse fidelement la courbe reguliere des Carpates depuis leur extremite meridionale jusqu'a celle N.O., pour se prolonger en suite a travers les Miuelgebirge de Boheme et d' Allemagne jusqu'en Westphalie. Les deux especes des extremites de la iigne (lateroproducta et thuringica) ne posent pas de probleme particulier. Mais les 4 chainons intermediaires posent des problemes. On remarque d'abord que (a l'exception de chomiacensis) elles habitent un territoire extremement restreint, les montagnes comprises entre les sources du Pmt et de la Tisa; trois especes coexistent dans les memes localites. Mais Ie cas Ie plus remarquable est celui de transsylvanica, espece actuellement connue par une seule population, et qui presente une variabilite tout a fait extraordinaire des genitalia 0' dont l'etude eclaire d'un jour nouveau tout Ie probleme des Annitella. L'etude du caractere Ie plus frappant des genitalia 0' (les prolongements des angles latero-apicaux du tergite VIII) montre que cette population presente un poly• morphisme accentue (Fig. 3): chez certains exemplaires l'aspect est celui de latero• producta, chez certains autres il est intermediaire entre cette espece et transsylvanica, et il y a aussi des exemplaires dont l'aspect correspond bien a celui du holotype de trans• sylvanica. La variabilite des autres pieces de l'armature 0' ne cOIncide qu'assezrarement avec celie des prolongements du tergite VIII. Heureusement, l'armature genitale de La 9, qui est caracteristique, nous apporte la preuve de La validite de l'espece transsylvanica (mais quelques femelles ont ete capturees in copula avec des 0'0' differant nettement entre eux!). Au polymorphisme phenotypique de transsylvanica correspond sans doute un poLymorphisme aussi accentue du genotype ('poLygenotypisme' selon BOESIGER). II faudra observer:

(a) que cette espece est pLacee a I'extremite N. de I'areal de lateroproducta et legerement au S. de l'areal des trois chainons qui font suite a transsylvanica; (b) que La variabiiite phenotypique du 0' ne se presente pas n'importe comment, mais qu'elle represente une oscillation entre Le type lateroproducta d'une part et Ie type dzicdzielewiczi-kosciuszkii-chomiacensis d'autre part. II est sur que La selection naturelle trouve dans la variabilite de ceUe popuLation un riche materiel sur Lequel elle peut operer et que [IOUS avons ici une illustration particulierement saisissante de l'idee que c'est la popuLation qui est Ie 'Laboratoire de l'Evolution'. Essaions maintenant de degager les enseignements que nous apporte Le cas des Annitella. Nous sommes ici devant une lignee orthogenetique indiscutabLe, dont tous les membres coexistent dans La nature actuelle ('orthogenese horizontaLe') et dont

61 Fig. 2 Distribution geographique des especes de la lignee orthogenetique des Annitella. La fleche marque Ie trajet emprunte par la lignee en expansion. I-triloba, 2-lateroproducta, 3-transsylvanica, 4-dziedzielewiczi, 5-kosciuszkii, 6-chomiacensis, 7-thuringica. 62 Fig. 3 Polymorphisme de I'armature genitale " chez l'unique population connue de Annitella transsylvanica. Bord distal du VIlle tergite vu de face, et aspect de I'angle latero• distal de ce tergite en vue laterale, chez un exemplaire 'du type lateroproducta' (a), chez un exemplaire intermediaire (b) et chez un exemplaire 'typique' de transsylvanica (c).

a

b

Fig. 4 Appareil penial en vue laterale chez 3 especes de Allogamus: uncatus (aJ, dacicus (b) et starmachi (c). 63 l'etude permet de suivre l'evolution par etapes de l'ensemble des caracteres genitaliens, depuis une forme visiblement plesiomorphe jusqu'a une forme nettement apomorphe et meme hyperthelique. On connait des cas du plus haut interet de series orthoevolutives dans la nature actuelle; Ie cas de certains isopodes terrestres de Turquie etudies par V ANDEL, celui de certains Homopteres Membracides etudies par BOULARD, en fin celui des coleopteres aquatiques du genre Dryops, etudie par STEFFAN. Le cas des Dryops s'apparente beaucoup a ceIui des Annitella, en ce qu'i! existe des correlations saisissantes entre Ie degre de differenciation des genitalia des deux lignees phyletiques reconnues par l'auteur allemand, et la distribution ecologique et geographique des especes. D'ailleurs, pour moi com me pour STEFFAN, il s'agit ici de cas d'orthoselection et d'ectogenese; personellement je n'ai jamais compris pourquoi l'influence du milieu et la selection devraient etre exclues de l'interpretation des series orthoevolutives. Mais les Annitella nous permettent d'aborder aussi d'autres problemes. L'origine de la lignee devrait etre cherchee en Peninsule des Balkans, pendant les glaciations du Pleistocene; la parfaite superposition de cette serie lineaire avec la chaine des Carpates et sa progression toujours vers Ie NO a travers les Mittelgebirge, prouve qu'elle a colonise ces montagnes au fur et a mesure de la retraite de la calotte glaciaire wiirmienne, les nouveaux territoires montagneux liberes etant colonises l'un apres l'autre, ce phenomene s'accompagnant des resultats de la speciation. Le laps de temps necessaire a la naissance de la serie de 6 especes lateroproducta• thuringica serait donc de 10000-13000 ans a peine, Ja lignee 'datee' des Annitella constituant donc une contribution concrete au probleme si peu connu de la vitesse (du tempo) de la speciation. Les Annitella nous mettent devant la speciation surprise sur Ie vivant et en plein deroulement: il me semble certain que transsylvanica est derivee de lateroproducta, que thuringica est issue de chomiacensis, et Ie cas de dziedzielewiczi-kosciuszkii-chomiacensis pourrait fournir, lorsqu'i! sera mieux connu, des preuves a l'appui de la possibilite de la speciation sympatrique. Les Annitella montrent de fa~on peremptoire que certains postulats de la theorie de HENNIG n'ont certainement pas de valeur generale. Et c'est en definitive surtout sur l'enorme importance des etudes populationnelles qu'attire l'attention Ie cas des Annitella. Je passe maintenant aux Allogamus carpatiques du voisinage de uncatus. L'ensem• ble de la chaine des Carpates est peuple par A. uncatus BRAUER, espece primitive et caracterisee par un penis rigide, en lame sclerotisee; c'est la seule espece avec ce type d'appareil penial qui habite les Carpates. Sur Ie fond de cet areal continu, 4 autres especes d' Allogamus, appartenant au meme groupe que uncatus, ont une distribution extremement limitee; elles se caracterisent, toutes les 4, par certains caracteres evidemment apomorphes, mais surtout par un penis qui est radicalement different de celui d'uncatus (Fig. 4): flexible, membraneux, probablement exsertile. Vne de ces especes, A. dacicus SCHMID, habite deux massifs des Carpates meridionales, present• ant une segregation saisonniere et altitudinale par rapport aux populations de uncatus

64 des memes massifs. Les 3 autres especes, decrites par SZCZESNY (A. starmachi, A. lazarei et A. tatricus) representent un cas absoluement remarquable: eUes sont connues uniquement d'une zone tres limitee des Tatra, ou eUes coexistent dans exactement les memes cours d'eau: deux ruisseaux a 1600 et 1680 m d'altitude; ces 3 especes sont fort voisines entre eUes et leur origine monophyletique est hors de doute. Il me semble que les conclusions suivantes s'imposent. Il s'agit, dans tous les cas, de produits de la speciation sympatrique. L'espece-mere est toujours uncatus, forme primitive, a large distribution, constittiee par un fort grand nombre de populations; c'est a partir de quelques populations de uncatus des Carpates meridionales que dacicus s'est individualise, fort vraisemblablement par intervention de mecanismes d'isolement ecologique; et c'est a partir d'un fort petit nombre de populations des Tatra de Pologne que s'individualiserent les 3 especes sympatriques de SZCZESNY, sans qu'intervienne aucune barriere geographique. Dans les deux cas, un caractere a ete visiblement favorise par la selection: c'est Ie penis membraneux, flexible, exser• tile, qui remplace dans ces 4 especes la lame rigide de uncatus (mais attention: je me refere a ce caractere phenotypique seulement comme au caractere Ie plus saillant qui caracterise les 4 types sur lesquels la selection s'est arretee). NuUement affectee par la naissance dans son sein de ces 4 especes, l'espece-mere uncatus continue d'exister par un nombre fort eleve de populations. La possibilite de la speciation sympatrique a souvent ete niee. PersoneUement, je pense que la speciation sympatrique doit jouer dans la nature un role beaucoup plus considerable que ne Ie pensent les adversaires de cette theorie, et je considere Ie cas des Allogamus du voisinage de uncatus comme particulierement significatif a cet egard. Pour moi, Ie support theorique principal de la possibilite et de l'importance de ce type de speciation me semble constitue par les recherches sur ce qu'on a appele 'polygenotypisrne des populations' (DOBZHANSKI, BOESIGER, etc.). Les Allogamus nous incitent aussi a quelques consideration sur la theorie de 'l'exclusion competitive' et sur les postulats de HENNIG concernant la cladogenese. L'existence dans la nature de tres nombreux faits d'exclusion competitive est une certitude, mais ce principe ne doit pas etre considere comme verite absolue (il n'y a apparemment pas d'exclusion competitive entre les 3 especes endemiques des Tatra de Pologne). Et, a propos de la theorie de HENNIG, Ie cas des Allogamus du voisinage de uncatus prouverait: qu'une espece parentale ne doit pas obligatoire• ment se scinder en deux especes filles, mais qu'eUe peut etre a l'origine d'un certain nombre d'especes, par plusieurs de ses populations; que I'espece parentale ne doit pas pour autant disparaitre, mais qu'elle peut parfaitement survivre, eventuellement meme sur Ie meme territoire que les especes-filles. Si les Annitella nous ont permis de reflechir surtout sur I'orthogenese et les Allogamus sur la speciation non-geographique, un autre groupe de Limnephilides europeens nous permettra de nous pencher un peu sur Ie mecanisme de l'evolution degenerative. Il s'agit du groupe si bien individualise represente par les genres Acrophylax et Chionophylax, dont les especes ont une biologie tres particuliere: eUes

65 ~a

Fig. 5 Aspect bouleverse des ailes anterieures et posterieures chez plusieurs exemplaires de Chionophylax monteryla.

66 montrent une preference marquee pour les eaux des zones situees au-dessus de la limite superieure de la foret, leur periode d'ec!osion imaginale correspond exacte• ment avec celie du debut de la fonte de la couverture de glace et de neige recouvrant lacs et cours d'eau, leur cycle de developpement est nettement determine par les conditions particulieres de leur habitat, et toutes les especes presentent certaines particularites generalement considerees comme 'adaptations au froid' (grande variabilite de la taille, tendance marquee au brachypterisme, etc.). Ce qui retient particulierement notre attention, c'est Ie cas de Chionophylax monteryla BOTS., espece du massif bulgare de Rila; ici, sur un fond de brachypterisme accuse (les ailes sont fort probablement non-fonctionnelles), on remarque chez plus de la moitie des tres nombreux exemplaires examines, de profondes anomalies de forme et de nervulation des ailes anterieures et posterieures (Fig. 5): apex echancre ou me me profondement dechiquete, depression de la partie apicale du bord costal de l'aile posterieure, nombreuses plaques sclerotisees formant comme des cicatrices sur la membrane, derangement plus ou moins profond de la nervulation soit provoque par les anomalies de forme des ailes, soit independant de celles-ci. Je me suis demande quelle peut bien etre la signification evolutive de cette remarquable variabilite, de ce grand nombre d'anomalies, qui m'avaient fait penser (bien entendu il s'agit la plutot d'une figure de style!) a une "epidemie de mutations faisant ravage au sein de cette population alpine de Trichopteres". Sachant que ce sont les specialistes de l'evolution regressive des cavernicoles qui sont les plus avances dans ce domaine, je me suis adresse a l'un d'eux, Ie Dr. H. WILKENS, en lui soumettant Ie cas de Chionophylax monteryla. Son interpretation des faits est la suivante: quand une structure devient non-fonctionnelle, elle montre une grande variabilite phenotypique, basee sur une variabilite genetique resultant du fait que les mutations degeneratives ne sont plus eliminees par la selection; Ie nombre des mutations degeneratives depassant largement celui des mutations constructives, les organes non-fonctionnels sont deteriores par la pression mutationnelle; c'est seule• ment a la fin d'un tel processus degeneratif que la variabilite pheRotypique et genotypique du rudiment diminue; les ailes de C. monteryla sont un exemple de structure en voie de degenerescence et qui est phylogenetiquement jeune (grande variabilite!); elles evolueront vers la disparition plus ou moins complete. J'avoue que cette explication me semble, grosso modo, satisfaisante. Je ne puis cependant m'empecher de me poser la question: pourquoi des especes si voisines de C. monteryla a tout points de vue (c. czarnohoricus DZIEDz. par exemple) ne presentent pas la me me profusion d'anomalies? Mais la question la plus troublante est la suivante: pourquoi la selection naturelle doit-elle etre exclue de ce processus? Pourquoi ne pas admettre que la selection opere sur les individus presentant des anomalies, en choisissant precisement ceux chez lesquels les anomalies sont les plus profondes, de sorte que Ie processus de degenerescence des ailes s'accentue et se generalise progressivement dans la population? Voici ce que je n'arrive pas a comprendre.

67 Bibliographie BOTOSANEANU, L. 1957. Une sous-espece balcanique brachyptere de Chionophylax czar• nohoricus DZIEDZ. - Beitr. Ent. 7: 598-603. BOULARD, M. 1973. Le pronotum des Membracides: camouflage selectionne ou orthogenese hyperthelique? Bull. Mus. Nat. Hist. Nat. (3) 109: 145-165. DOBZHANSKI, Th. & BOESIGER, E. 1968. Essais sur l'Evolution. Masson & Cie., Paris. MURGOCI, A. & BOTOSANEANU, L. 1957. Genul Annitella KLAP. in R.P.R. Analele Univ. Bucure~ti, Ser. sti. nat 13: 139-149. SCHMID, F. 1951. Notes sur quelques Halesus. Bull. Soc. Vaud. Sc. Nat. 65: 63-71. --. 1952. Le groupe de Chaetopteryx. Rev. Suisse Zool. 59: 99-171. STEFFAN, A. W. 1961. Autogenese oder Orthoselektion in der Evolution der Dryops-Arten (Dryopidae, Coleoptera)? Naturwiss. 48: 28. SZCZESNY, B. 1967. Notes sur quelques especes d'Allogamus (Trichoptera, Limnephilidae) dans les Tatra. Bull. Acad. Pol. Sci., Ser. sci. bioI. 15: 479-482. VAN DEL, A. 1973. L'utilisation des isopodes terrestres (Oniscoides) en tant qu'indicateurs des modalites de l'evolution animale. c.R. Acad. Sci. Paris 277: 1233-1236. WILKENS, H. 1971. Genetic interpretation of regressive evolutionary processes: studies on hybrid eyes of two Astyanax cave populations (Characidae, Pisces). Evolution 25: 530-544.

Discussion

ILUES: It is very important to say that this statement is opposite to the HENNIG theses. It is the normal idea of speciation that one species branches up into two. This gives the basis of HENNIG'S idea of apomorphic and plesiomorphic groups. All that is true to my opinion, and it will not be less true after this presentation. As we have a mother-species with 3 or 4 descendents, it seems to be an exploding group. I think we can still stick to the general idea of the branching into a plesiomorphic and an apomorphic species. This here is as far as I know the first rather sure case of poly group species-radiation of a mother-species. VAILLANT: We can imagine that one mother-species gives free at the same time at different places some daughter species, and all come together a little later. MORSE: I think that your idea about sympatric speciation is very interesting and I should like to know what you think may be the mechanism by which the population initially became reproductively segregated in order to initiate the process of speciation. BOTOSANEANU: You will find a very fine explanation of this mechanism when you will read especially the publications of DOBZHANSKI and the papers of his pupil, BOESIGER. They have found that is not at all true that natural selection tends to make animal populations genetically uniform. They found that, at the contrary, natural selection tends to ensure genetical and also phenotypical diversification inside each animal population. This phenomenon was called polygenotypism. They never have used their findings as basis for the presence of sympatric speciation in nature. This problem, I don't know why, was not interesting for them. But I think that the discovery of the tendency toward polygenotypism, toward genetical diversification inside an animal population, offers a clue for understanding why not only geographic speciation does exist in the nature. MORSE: Yes, we know that there is diversification within almost any natural population, but all members of this population nevertheless interbreed, presumably. How do you suppose that the interbreeding breaks down? BOTOSANEANU: You can read some fine papers by DOBZHANSKI and BOESIGER and certainly by other genetists working on Drosophila (willistoni and paulistorum for instance), and you will find that inside the population there are instances in which not all members of a population

68 interbreed at random. There is a selection, isolation, there is some preference of some individuals for some particular mates. A good genetical study of Annitella transsylvanica will probably show a similar phenomenon, I suppose. Unfortunately our Trichoptera were almost never used for studies of evolutionary genetics. HILEY: What about the larvae of those sympatric Allogamus? BOTOSANEANU: Elles ne sont pas connues. Peut-etre M. SZCZESNY possede des larves.

HILEY: Is it possible that these larvae live in quite different habitats within the stream~ BOTOSANEANU: We know nothing about it, but it is possible. When I said that the three ,pecies are quite sympatric this can be somewhat wrong. Perhaps the habitat is not quite the same. The streams are the same. Maybe there is an ecological segregation. STATZNER: I think the phenomenon of polygenotypism you spoke about can become recogniz• able in the difference of the genital structures of specimens of one population. That may cause an isolation within one population, and may be an important mechanism for the origin of sympatric species. Results found in African Cheumatopsyche justify that speculation. BOTOSANEANU: So you mean that in Cheumatopsyche it can be the same? Yes, I agree with your opinion that the 'SchloB-Schliissel-Prinzip' is a valid one. Ross: I will like to point out that most of these different species of Drosophila paulistorum are now sympatric but it is DOBZHANSKI's view that they evolved originally as allopatric species whose ranges later spread and became sympatric. BOTOSANEANU: You express a similar opinion in a fine paper you have written on some Allocapnia. I have read this paper with much interest, but I wonder why do you not accept the sympatric origin of these Allocapnia. I don't know why, instead of being of this opinion, you describe a rather complicated system of migrations, of species separation by glaciation and post-glaciary events and than meeting of some formerly separated species. For me your paper is a very fine example of sympatric speciation Ross: But not for me. In my view it is a marvellous example of allopatric speciation. I believe the difference in our view is that in eastern North America we have a fairly good knowledge of the Pleistocene geologic and climatic histories, and our Allocapnia conclusions follow these events very closely. I don't know the intimate geology of your Carpathian area. BOTOSANEANU: It's fairly well known. The Pleistocene events are fairly well known. Ross: When you realize that we had a series of wet cool climates interspersed with warm dry climates, you have all the mechanisms that you need for first isolation of two sister populations, and then during warm dry periods the evolution of these separated populations into distinct species. BOTOSANEANU: I am not of the opinion that sympatric speciation must replace all other ones. The geographic speciation is the most important mechanism. But there are also such cases. Ross: In my recent book on biological systematics I pointed out several instances in insects in which sympatric speciation without doubt did occur. The groups involved were sawflies and leafhoppers. NIELSEN: A partial second generation resulting in two separate flying seasons in a species, as has been shown in Cheumatopsyche, might provide a means of sympatric segregation.

FLINT: Dr. RESH shows in his paper a case that might produce sympatric speciation. He shows

69 two cohorts emerging at different times of the year, but each cohort might have a full year cycle, but emerging at different times. If they emerge at different enough times, there might be no interbreeding and in time speciation. Ross: An extremely cogent point. This type of staggered life histories is one of the most important mechanisms for sympatric speciation in insects, that is, having two cohorts in which the adults of one cohort come out and interbreed before the adults of the other cohort come out to breed. This model fits sympatric speciation very nicely and I think we have good phylogenetic evidence in the sawflies that this has indeed happened. MALICKY: From where do you derive the information that these three species of Allogamus in the Polish Carpathians are originally derived from A. uncatus? Could it also be in reverse? BOTOSANEANU: Allogamus uncatus has thousands of populations, and the three species are extremely isolated, represented by only a very low number of populations. MALICKY: I could imagine that these rare species, not only those in Poland, but also the others in other countries (dacicus etc.), have a common ancestor which is now extinct, and they may now have relict populations, and uncatus distributed independently. Of course, they belong to the group of uncatus, but uncatus is a rather young species which may have evolved later and has now a very wide distribution area. What would you say about this? BOTOSANEANU: The three Polish species cannot be relictary, because they were not found, for instance, in the Middle East in some semidesertic area, but in the Carpathians! In the southern Carpathian species, dacicus, there is a fine altitudinal segregation: uncatus only till to the superior limit of forest, and dacicus (which in my opinion is derived from uncatus) only above the timber line, above 2000 m. MALICKY: As far as I know, uncatus is the only species of this group which has an internal hook on the inferior appendages (to be seen in caudal view). In addition it is also (except A. mendax and A. stadleri from Southwestern Europe) the only of this group with a heavily sclerotized aedeagus. The others (dacicus, starmachi, tatricus, lazarei) have a soft aedeagus. If the three Polish species are really offsprings of uncatus, these two characters must have been evolved convergentIy in these and in dacicus. BOTOSANEANU: Yes. I have forgotten to emphasize the important role played by natural selection: this character (non-sclerotized phallus) was obviously selected by natural selection, being certainly an apomorphous feature as compared with the heavily sclerotized phallus of uncatus. MORSE: I am not familiar with the European species of this group, but all these character states (e.g. the hook on the gonopod and the sclerotized phallus) which also appear in closely related groups are probably the primitive condition for this group. Other character states are thus derived in the group. By this logical process it is possible to determine the direction of evolution. You may thus propose the direction of evolution for the group without relying on much more tennuous geographical and ecological courses of reasoning.

70 Proc. of the First Int. Symp. on Trichoptera, 1974, Junk, The Hague

A progress report on studies on Trichoptera of the Eastern Mediterranean Islands*

HANS MALICKY

The investigation programme which started in 1971, concerns the ecology, systema• tics and zoogeography of the Trichoptera of the Mediterranean region. Field work was carried out in Crete (April-May 1971, September-October 1972, July-August 1974), Cyprus (April-May 1974), Andros (May 1973), Euboea (May 1974), and on each of these occasions field studies were also made in several mountainous regions of Peninsular Greece. Colleagues also made available material from the islands of Thasos, Limnos, Samothraki, Ikaria, Chios, Rhodos and Karpathos. In continuation of the programme, trips are planned to Libya, the Baleares, and to as many as possible of the islands of the Aegean. Most promising in this respect are Naxos, Paros, Tinos, Samos, Chios, Kos and Lesbos. The field methods include collection of adults with nets and light traps, aquatic stages with nets and bottom-samplers, temperature recording, chemical analysis of water, and photographic documentation of the biotopes. The preliminary results are summarized below.

1. Systematics The Trichoptera fauna of the islands is rich and contains a good number of endemics. The recorded fauna of islands where I collected is shown on the table on the following page. A total of approximately 150 species is known to me from peninsular Greece, the Greek islands, and Cyprus; about 40 of them are new to science. The genus Tinodes (Psychomyidae) has an unusually high number of species in the Aegean region; 14 species have been found up to date, of which not less than 12 are new to science. More new species of Tinodes are expected from other islands. About one-third of the European Tinodes fauna consist of endemics from the Aegean region. About 20 to 30 further new species of Trichoptera can be expected to be found in the region. 2. Ecology

The limnephilid species of Micropterna, Stenophylax and Mesophylax are highly adapted to conditions in streams which dry up in summer. The adults are known to aestivate in caves in Central Europe, but not a single specimen has been found up to

* Mit Unterstiitzung des Fonds zur Forderung der wissenschaftlichen Forschung in Osterreich

71 (Species number)

Crete Cyprus Euboea Andros

Rhyacophila 1 1 3 1 Agapetus 1 1 1 1 Hydroptilidae 8 5 1 1 Philopotamidae 5 1 5 2 Hydropsychidae 1 4 2 2 Psychomyidae 4 2 2 4 Micrasema 1 Lirnnephilidae: Microptema group 7 3 5 Limnephilus 1 2 1 Apatania 1 Thremma 1 Leptoceridae 1 2 2 Sericostomatidae 1 2 1 Lepidostomatidae 1 1 Beraeidae 2 1 2 1 HeJicopsychidae 1 Calamoceratidae 1 1 date in caves in Crete, despite considerable efforts. I presume that in Crete they aestivate in soil fissures etc. The very vagile adults can easily recolonize streams in which the local population has died out because of drying up under extreme conditions. Their larvae show considerable individual differences in rate of develop• ment. These features are adaptations to the conditions in intermittent streams. Most other Trichoptera are confined to the few permanent streams which have a maximum temperature of around 20°C in summer when the air temperature exceeds 30°C. The larvae of Limnephilus minos live in brackish water of low concentration (2%0) and a constant temperature of 15-16°C in the intertidal region. Stagnant water bodies are lacking except for residual pools in rocky stream beds, artificial ponds, and one lake. Residual pools and artificial ponds have a rich fauna of aquatic insects but no Trichoptera. It is suggested that the absence of Trichoptera larvae results from high summer temperatures and oxygen deficit. In Lake Kumas only one species, Hydroptila kumas, has been found in numbers; this species also lives in streams. The species typical of ponds and lakes in northern and central Europe are absent. In the streams of Crete and Andros no longitudinal zonation has been found. The potamal (in the sense of ILLIES) is not found in the islands because of the length of streams which do not usually exceed 10 km; only the leropotamos has a length of about 30 km. I have no explanation for the lack of the krenal. Springs are inhabited by an impoverished stream fauna. It should be noted that in Crete no vertical zonation has been found in the terrestrial vegetation. In Cyprus, on the other hand, three longitudinal zones may be distinguished in the streams: (1) a spring-brook zone 72 with cold water of approx. 9°C at about 1600 m in the Troodos Mts.; (2) an intermediate zone between from about 1500 m down to 900 m; and (3) a zone below about 800 m. Only the intermediate zone has a rich fauna. The spring-brooks are inhabited by very few insects but in high abundance; Plectrocnemia renetta is the only trichopteran. The lowland zone comprises only hydroptilids, Agapetus caucasicus and Polycentropus milikuri. 3. Zoogeography The Trichoptera of the European fauna may be divided into four groups: (A) Inhabitants of permanent streams with a wide distribution, and without much zoogeographical significance. (B) Species of presumably late Tertiary or interglacial origin. The differentiation into species of local distribution may be intensive, as for example in the genus Tinodes which has many species in the Aegean region. It is only surpassed by the coleopteran genus Hydraena with about 60 species there. (C) Young immigrants. Perhaps the whole family Limnephilidae belongs to this group, but also for example the species group around Hydropsyche instabilis. They are well adapted to recent eco,ogical conditions and have vagile adults. In the Mediterranean region and in the mountain systems of central and southern Europe they may have either a wide distribution, or a tendency to split into local species or races. The former is evident in the Hydropsyche instabilis-group and in the limnephilid genera Drusus, Annitella and Apatania. (D) Endemics which have relict areas and whose closest relatives are geographically well separated, usually by the sea, over a long geological time. They live in permanent streams. Several examples are given in Fig. 1. Up to the present I know about a dozen pairs or groups of species which fit into this scheme. They belong to the genera Rhyacophila, Agapetus, Pti loco lepus, Polycentropus, Plectrocnemia, Paduniella, Hydropsyche (subgenus Caldra), Silo, Thremma, Beraeamyia, Helicopsyche, Notidobia, Odontocerum and Calamoceras. Species of this distribution type are frequent in the islands. They are probably relicts of a Mediterranean continent of early Tertiary (Eocene?) origin. The existence of species of this very great age (about 30-40 millions of years) is plausible, because in certain regions of southern Europe there have been no essential changes in the ecology of mountain streams since that time. The family and genus composition of the fauna is similar to that of the Baltic amber from the Eocene, except for the important difference that the amber fauna contains a large number of very abundant stagnicolous polycentropodid species which must have disappeared later due to climatic changes. If this hypothesis is accepted, the study of Mediterranean Trichop• tera offers important contributions to the zoogeography of Europe as a whole. On the other hand, the presence of species of this distribution type demonstrates clearly that these islands were never completely submerged by the sea after the early Tertiary. This may be true, for example in Cyprus, because of the presence of Rhyacophila aphrodite. The faunal connections, that is the proportions in the number of species belonging to the above mentioned groups, differ from island to another. Comparing Crete, 73 -...l ~

~

o <><:0

Fig. 1. Distribution of three groups of Trichoptera of palaeo-mediterranean disjunctibn type (simplified): Circles: Rhyacophila trifasciata group; from left to right: R. angelieri DECAMPS (Pyrenees), R. trifasciata MOSELY (Corsica, with R. pallida MOSELY and R. tarda GIUDICELLI), R. rougemonti MeL. (Southern Italy), R. gudrunae MALICKY (Crete), R. aphrodite MALICKY (Cyprus). Triangles: Genus Calamoceras. Iberian Peninsula: C. marsupus BRAUER, Aegean region: C. iIliesi MALICKY & KUMANSKI. Squares: Genus Beraeamyia. Southern France : B. squamosa MOSELY. Balkan Peninsula, from top to bottom: B. hrabei MAYER, B. schmidi BOTOSANEANU, B. kutsaftikii MALICKY (Euboea), B. aphyrte MALICKY (Crete). Cyprus and Euboea, the percentage of younger immigrants (Micropterna, Hydrop• syche) is about the same. The proportion of species of palaeo-mediterranean distribution type is higher in Crete and Euboea than in Cyprus. The same is true for the inhabitants of permanent streams which have large areas. Euboea shows strong faunal connections with peninsular Greece and the Balkan Peninsula, while Crete and Cyprus have only very slight ones. The connections of Asia Minor and the Caucasus are striking in Cyprus, weak in Crete, and absent in Euboea. Cyprus also shows slight faunal connections with Palestine, as demonstrated by the presence of Tinodes kadiellus and T. negevianus which have been described from Israel. The survey of the other islands is not yet good enough to demonstrate such conclusions, but even the limited material shows strong connections between the eastern Aegean islands and Asia Minor.

Discussion

VAILLANT: Did you not find one single species in North Africa? MAUCKY: I have much material from the western part of Northern Africa, but it contains no species of this palaeo-mediterranean distribution type. I should also say that I got material from Aspromonte (in the extreme south of the Italian Peninsula), which is very similar to the fauna of the Alps. ILUES: First of all I want to congratulate you on these very interesting findings. It is a good step forward in understanding of European history of freshwater insects. May I offer you an explanation for the lacking of zonation in Crctc~ We had the same case occurring in New Guinea, there is no zonation at all, we have no cool-adapted forms. If they once have become extinct for some reason, immediately the resting species climb up the hill in the gaps which are open now after the extinction of the cool-adapted groups. It would suffice one warm day in all the millions of years to get extinct all the cool-adapted forms! There may have been some warming up for a certain time with the consequency of the extinction of cool-adapted groups. If we can really compare now the islands of Cyprus and Crete which are in the same latitude, and you state there is zonation in Cyprus and no zonation in Crete, we can draw the conclusion that Cyprus in all its history had places for cool-adapted groups, and Crete at some time or another had lost the biotopes for cool-adapted groups. There are no means of re-immigrating. MAUCKY: I think that cool-adapted groups had never reached these islands, at least Crete and Cyprus (with the possible exception of Apatania cypria in Cyprus). ILUES: But what is responsible for the zonation in Cyprus? MAUCKY: In the high mountain springs in Cyprus in cool water lives Plectrocnemia renetta (which is also known from the island Ikaria), a species of old-mediterranean origin. Its closest relatives were found in Euboea, in Crete, and in Southern France. P. kydon from Crete and Euboea is not involved in the zonation. But the problem is still more complicated. For instance Hydropsyche discreta is found everywhere in Crete, from the high mountain springs (if present at all) to sea level. But in Cyprus it (or a close relative - this is still unclear) is found only in the intermediate zone. It was impossible to find Hydropsyche in the lowland zone in Cyprus. It seems also that Plectrocnemia renetta in Ikaria is not involved in the zonation. Another instance: The river crab, Potamon potamios (however in other subspecies), is common in the intermediate zone in Cyprus, and in Crete it is everywhere. ILUES: So we better would not use the word of "zonation", if it's just for one species. I think that this is not the type of liIpnological zonation.

75 MALICKY: These are only examples. There are more such instances. ILLIES: What about the Calamoceras? MALICKY: One species has been well-known for a long time. C. marsupus was described from Gibraltar by BRAUER from the material of the Novara expedition, but it occurs also in Portugal, in Central Spain, and, according to the literature, it is found in Southern France. The other one, C. illiesi, lives in Andros, Euboea, and in the Strandscha Mountains in Bulgaria. In addition MARTYNOV described one species from the Caucasus but did not name it because he had females only. In Andros and in Euboea it is the usual stream of the not-zonated type in which Calamoceras illiesi lives. FLINT: Recently, I have been reading about the probability that the Mediterranean dried up. Do you know the age of this? MALICKY: The time of drying up of the Mediterranean according to the literature was much later, in Pliocene. The connections of the caddis-flies must be much older. FLINT: When the area was dry, or nearly dry, don't you think that the streams must have dried also? MALICKY: Only a few streams must have had water all the time. If they had been completely dry for only one year during this long time, the species would not exist now. By this time the species were already confined to the islands where they live now. CRICHTON: Have you any information about the state of maturity of females of limnephilids, and on aestivation? MALICKY: As concerns the Stenophylax group it is well-known. The females and males aestivate, and the development of the gonads starts only in late summer. As concerns Limnephilus minos I am not yet clear about the phenology. Maybe it has two generations and direct development. I can only say that I have adults from April-May and October, and I found full-grown larvae in October, and half-grown larvae in July and August. MORETII: Le Limnephilus minos n'a pas ete trouve dans les eaux douces? MALICKY: Non. J'ai trouve les larves dans deux localites en Crete, dont l'eau est un peu saumatre, mais la concentration est seulement 2%0. Ce n'est pas beaucoup. VAILLANT: What sorts of places are the Tinodes larvae living in? Not hygropetric? MALICKY: In Crete T. reisseri and T. aligi usually live in permanent streams, but they may also occur in hygropetric localities. But I have the impression that in the genus Tinodes only a few species are strictly confined to hygropetric localities, e.g. Tinodes zelleri in the Alps, and T. raina in the Peloponnesos. VAILLANT: Did you find any Stactobia? MALICKY: Yes, Stactobia caspersi. It has been described from Bulgaria and occurs also in the mountains of Central Greece. I found it in the Peloponnesos and in Crete. In addition, S. mounioti has been described from Cyprus, but only as a larva, so it cannot be said whether it is the same or not. MORETII: Vous avez aussi trouve Enoicyla? MALICKY: Oui, j'ai trouve E. costae en Grece du Sud. Cette espece n'est pas rare, mais on doit faire les collections en Octobre. VAILLANT: Pas d'Helicopsyche? MALICKY: Yes, Helicopsyche megalochari in Andros, whose closest relatives are H. sperata from Italy, H. lusitanica from Portugal, and H. revelieri from Corsica. H. bacescui from Romania is not close to these four. 76 Proc. of the First Int. Symp. on Trichoptera, 1974. Junk, The Hague

The differentiation of Drusus species of the group bosnicus

MARA MARINKOVIC-GOSPODNETIC

Abstract

As differences in the appendices intermediales and appendices superiores of populations of Drusus group bosnicus from various parts of Bosnia had been noticed, the problem of their differentiation has been studied. At the same time attention has been paid to the distribution of these populations.

Introduction

KLAPALEK (1900) described Drusus bosnicus giving as localities the springs of the river Bosna, of the river Krupa (a tributary of Zujevina) and river Zujevina near Pazaric. RAoOVANOVIC, (1942) described Drusus plicatus from western Macedonia (Yugoslavia). SCHMID (1956) established the group bosnicus composed of Drusus plicatus RADOVANOVIC Drusus bosnicus KLAP., Drusus nigrescens MEYER-DUR and Drusus monticola McL. MARINKOVIC-GOSPODNETIC (1970) found in Bosnia and Herzegovina three new species from the group bosnicus: Drusus klapaleki, Drusus radovanovici and Drusus ramae. Research of MARINKOVIC• GOSPODNETIC (1971) showed that in Bosnia and Herzegovina, beside the species of group bosnicus, there are some other endemic species of the genus Drusus. The species of the group bosnicus, however, are most common and their areas are usually small. Some populations or the groups of populations could not be classified among the species that have been described: they keep apart by the structure of male genitalia. Nevertheless, these populations as well as the species of the group bosnicus are close. It was, therefore, worth while to study their differentiation.

Results

Drusus species of the group bosnicus inhabiting the mountain streams of Yugoslavia can be classified in two groups according to male genitalia: the first group (D. plicatus, D. radovanovici, D. ramae) looking fairly primitive, with simple appendices intermediales (Fig. 1); the second one (D. bosnicus, D. klapaleki), more evolved, with complex appendices intermediales which bear a characteristic process (Fig. 2). 77 In the species D. bosnicus these processes are very massive, in the form of wide wings, particularly pronounced in the specimens having processes darker than the base of appendices intermediales. In D. klapaleki the processes are short, narrow, with pointed tip turned upwards (Fig. 3). The areas of two groups are apart (Fig. 4). The first group inhabits the western Dinarids, the streams of Adriatic and Black Sea drainage systems; the area of the second group is on east of the area of the first group (only the population in the source of the river Ribnik is on the west), comprising the streams of the Black Sea only. Among populations with primitive appendices intermediales, two groups of popu• lations are different from the species that have been known. One group inhabits the springs' in Livanjsko polje, the spring of the river Pliva and a spring on the left bank of the river Vrbas, near Bocac. This group is fairly similar to the species D. radovanovici, being separated from it by the area of D. ramae. The second group of populations lives in the left tributaries of the river Bosna (Zujevina, Fojnica, Lasva) and in some right tributaries of river Vrbas (Bistrica, Kruscica). Among populations with complex appendices intermediales, only one population distinct from the species that has been described has been found. It lives in the spring of the river Ribnik (a tributary of the river Sana). It looks to me that Drusus populations from Livanjsko polje, from the river Pliva and source on the left bank of the river Vrbas are a new subspecies of D. radovanovici, while the populations from the left tributaries of the river Bosna and from the right tributaries of the river Vrbas are a new species, as well as the population of the river Ribnik.

Drusus radovanoyici septentrionis ssp.n. (Fig. 1d)

MALE GENITALIA. The difference between D. radovanovici radovanovici and. D. radovanovici septentrionis appears in the form of appendices intermediales and appendices superiores as well as in the shape of tubercules zone of the 8th tergite. In lateral view the appendices intermediales of D. radovanovici septentrionis are very massive, cut up posteriorly, the apex on the dorsal part is blunt and triangular. The posterior end of the appendices superiores is concave, their ventral part is little longer than the dorsal part. The tubercules of the 8th tergite form two large semicircular zones which are in contact along the median line. A part along the median line and along the posterior end is beset with a great number of tubercules and it is darker. The light field with few tubercules is small, entering slightly between two dark semicircular zones. Drusus radovanovici septentrionis inhabits the north-western part of Bosnia. The localities are: -the source of the river Pliva, 130 2S?, 8.6.72; 100 4S?, 4.5.73. (Holotype 0, allotype S?, paratypes 00S?S? are in author's collection, the Faculty of Natural Science and Mathematics, Sarajevo.)

78 a

I'

a ~b ---." c --' - e Fig. 1. Male genitalia of Drusus species of the group bosnicus with simple appendices intermediales: (a) D. plicatus; (b) D. -.l radovanovici radovanovici; (c) D. ramae; (d) D. radovanoviCi septentrionis; (e) D. medianus. \Q -the sources of the rivers in Livanjsko polje: river Bistrica, 70, 8.6.66; 90 1<;', 19.10.70; 10 1<;', 16.7.71; Sturba, 20 2<;', 8.6.66; 290 19<;', 17.10.70; 40 5<;', 15.7.71. -the source on the left bank of the river Vrbas near Bocac, 100,2.6.1966; 40 4<;', 8.6.1972.

Drusus medianus sp.n. (Fig. Ie)

MALE GENITALIA. Appendices intermediales differ from those of other Bosnian species of Drusus gr. bosnicus in being small. They bear two narrow tips on the dorsal part. In dorsal view, appendices intermediales are narrow and with a deep recess laterally. Appendices superiores are concave and similar to those of D. radovanovici septentrionis. The surface beset with turbercules is also similar to that of D. radovanovici septentrionis, but its dark semicircular zones are more separated by the light zone beset with few tubercules. Localities in the springs of the left tributaries of the river Bosna are: -Lasva (Plave vode), 300 12<;>,5.6.1965; 80 4<;', 4.5.1973. Holotype 0, allotype <;', paratypes 00<;'<;' in author's collection. -Lasva (Komar), 120 4<;', 13.5.1971; 60, 4.5.72; 40 3<;', 15.5.73. -Zujevina, 50 2<;', 10.5.70. -Zujevina (Ljubovcica potok), 260 2<;',14.5.1957; 501<;',15.5.1958; 20 23.4.1958; 120 3<;', 15.5.1960. -Zujevina (Krupa), 20, 28.6. 1957. -Fojnica (Pozama), 220 10<;', 31.5.1967. Localities in the springs of the right tributaries of the river Vrbas are: - Bistrica (near Gomji Vakuf), 10, 15.6.1973. - Kruscica (tributary of the river Bistrica), 10, 18.6.72.

Drusus vespertinus sp.n. (Fig. 2c)

MALE GENITALIA. The structure of appendices intermediales is complex. On the posterior end they bear two fingerlike processes directed backwards, in lateral view, and laterally, in dorsal view as wall as in posterior view. The processes are much lighter than other part "base" of appendices intermediales. The shape of the dark part of appendices intermediales is very similar to appendices intermediales of D. radovanovici septentrionis. Appendices superiores are concave and similar to those of D. radovanovici septentrionis and D. median us. The zone beset with tubercules is fairly large. It is in the shape of two dark semicircular fields, separated by a light field beset with few tubercules. Density of tubercules of the dark fields is greater at its posterior end, which is darker and again in the form of two smaller semicircular fields. D. vespertinus has been found only in the large karstic spring of the river Ribnik, a tributary of the river Sana: 680 5<;', 26.3.1968 (Holotype 0, allotype <;', paratypes oon are in author's collection); 200,11<;',25.5.1968; 120 1<;',28.3.1975. 80 u

Fig. 2. Male genitalia of Drusus species of the group bosnicus with complex appendices intermediales: (a) D. bosnicus; (b) D. klapaleki; (c) D. vespertinus.

81 a

A

I :" I:. A it.•' .!~ K-: ~. a b c Jtd e B ~ )( ][ f 9 h Fig. 3. Appendices intermediales (and appendices superiores) of Drusus species of the group bosnicus (A-lateral view, B-

Discussion

Although the structure of female genitalia has not yet been completely studied, it seems that there are no big differences among species of Drusus gr. bosnicus that inhabit Yugoslavia. At the same time, the structure of male genitalia shows the difference as well as great relationship of all species. Particularly interesting is the great similarity between some neighbour species with simple appendices inter• mediales and with complex appendices intermediales, e.g. D. radovanovici septen• trion is and D. vespertinus. Similarity may be seen not only in appendices inferiores and appendices superiores but also in appendices intermediales. Appendices inter• mediales of these species differ by presence or abscence of two processes on the

82 Fig. 4. Distribution of Drusus species of the group bosnicus in Yugoslavia: species with simple appendices intermediales (striped) and species with complex appendices intermediales (dotted). posterior end. It is also the case with D. radovanovici radovanovici and D. klapaleki, D. ramae and D. bosnicus. (Fig. 3). Appendices superiores have the same shape in the species with simple appendices intermediales and species with complex appendices intermediales. It should be noticed that species with tightly close areas on northwest (D. vespertinus, D. median us and D. radovanovici septentrionis) respectively the neighbour species on southeast (D. klapaleki and D. radovanovici radovanovici) possess the same form of appendices superiores, while their appendices intermediales may be similar or not. All species of Drusus gr. bosnicus are allopatric forms (Fig. 5). It is interesting that areas of some species with simple appendices intermediales and of some species with complex appendices intermediales are tightly connected. It is particularly the case

83 Fig. 5. Distribution of Drusus species of the group bosnicus in Yugoslavia: (A) D. vespertinus; (B) D. radovanovici septentrionis; (C) D. medianus; (D) D. ramae; (E) D. bosnicus; (F) D . klapaleki; (G) D. radovanovici radovanovici; (H) D. plicatus. with D. radovanovici radovanovici and D. klapaleki in the drainage system of the river Sutjeska. Theyoccuron some places which maybe only afew kilometers apart. Taking into account male genitalia and distribution of species, it may be concluded that all species of Drusus gr. bosnicus are related. It seems that the species with simple appendices intermediales are ancestral to species with complex appendices inter• mediales. Among the northwest populations with simple appendices interniediales, the species D. vespertinus has been developed and among the southeast populations the species D. klapaleki with complex appendices intermediales. The process of differentiation of new species is pronounced particularly in change of appendices intermediales. Appendices superiores, however, remain unchanged and, to some extent, they are witness of closeness of these species.

84 The detailed studies of male genitalia and of species distribution have indicated not only the process of speciation but also the differentiation among populations of the same species. That is a problem of our further research.

Summary Two new species and one new subspecies of the genus Drusus gr. bosnicus (D. vespertinus, D. medianus, D. radovanovici septentrionis) have been described. There are two specie~ groups: a group with simple appendices intermediales (D. plicatus, D. radovanovici radovanovici, D. ramae, D. radovanovici septentrionis, D. medianus) and a group with complex appendices intermediales (D. klapaleki, D. bosnicus, D. vespertinus). Existing difference in male genitalia, however, does not cancel the impression of closeness of all species, even of the species from the first and the second group. It has been suggested that the species having simple appendices intermediales are ancestral forms to the species with complex appendices inter• mediales. It has been shown that the species with simple appendices intermediales has given origin to D. vespertinus on northwest, and to D. klapaleki on southeast.

References

KLAPALEK, F. 1900. Beitrage zur Kenntniss der Trichopteren- und Neuropterenfauna von Bosnien und Hercegovina. Wiss. Mitt. Bosn. -Herceg. 7; 671-682. MARINKOYIC-GOSPODNETIC, M. 1970. Descriptions of some species of Trichoptera from Yugo• slavia. God. BioI. inst., Sarajevo, 23: 77-84. --. 1971. The species of the genus Drusus in Yugoslavia. God. BioI. inst. Sarajevo, 24: 105-109. RADOYANOYIC, M. 1942. Uber zwei neue Trichopteren-Arten aus Mazedonien. Zool. Anz. 140: 183-190. SCHMID, F. 1956. La sous-famille des Drusinae (Trichoptera, Limnophilidae). Mem. Inst. Roy. Sci. Nat. Belg. 2. serie 55: 1-92.

Discussion

CIANFICCONI: Are there scales or setae in the posterior wings of Drusus bosnicus? MARINKOYIC-GOSPODNETIC: They possess setae. FLINT: Did you find any of these populations that were limited to a single drainage area? MARINKOYIC-GOSPODNETIC: Some species are not only limited to a drainage area but even to some tributaries of a river: D. bosnicus inhabits only the right tributaries of the river Bosna and D. ramae inhabits only the right tributaries of the river Neretva. NEBOISS: What would be approximately the average height of the mountains, of the watershed? MARINKOYIC-GOSPODNETIC: It is between 1500 and 1900.

85 Proc. of the First Int. Symp. on Trichoptera, 1974, Junk, The Hague

Some informations on the orobiontic fauna of Trichoptera of the Italian Western Alps above 2000 m 1

G. P. MORETTI, A. VIGANO & M. I. VIGANO-TATICCHI

During the years 1964 and 1965 investigations on aquatic stages and adults of Trichoptera have been made in the Alps between Monte Rosa and Maritime Alps in 122 stations between 250 and 2700 m above sea level. Except of three, these stations are on Italian territory. Field research has been carried out in August, September, October and November of 1964 and in June and August of 1965. Various kinds of water bodies, such as springs, waterfalls, brooks, torrents, lakes, ponds, marshes and peat-bogs, were investigated. Many of the 122 stations have been examined more than once, and in several of them sampling has been effectuated in order to ascertain the environmental factors which might influence the aquatic stages (temperature, pH, hardness, dissolved oxygen" depth, stream velocity and composition of the bottom). Samples have been taken during daytime and also by night with the aid of fluorescent lamps. The stations may be subdivided as follows: 31 stations below 1000 m, 70 stations between 1000 and 2000 m, 21 stations above 2000 m. The study of the gathered material is not yet completed. Actually we can only give a table of orobiontic fauna, concerning the Trichoptera found in the 21 stations situated between 2000 and 2700 m (Fig. 1). We just group the biotopes according to their ecological peculiarities (Fig. 2). Two stations (No. 17, 18) are morainic pools with a slimy and stony bottom and with not hard, oxygen oversaturated and sufficiently acid waters, subject to drying up in summer and freezing in winter; seven stations (No. 20, 28, 28A, 28B, 29, 63, 65) are small alpine lakes with clear, slightly acid or neutral waters; the hardness is low (except in No. 20) and with high oxygen values, generally higher than V.S., in spite of the presence of springs (min. 81 %, max. 143%). The remaining stations are lotic

1 This paper is a part of the investigation program on the trichopterofauna of the Western Alps, financed by the Italian C.N.R. We take the opportunity for thanking Prof. ATHOS GOIDANICH, Director of the Entomological Institute of the University of Torino and of the Centro di Studio di Entomologia Alpina e Forestale del C.N.R., for having entrusted us with the present research.

87 S E S \ r-\~ " i, \

Fig. 1. Sampling stations in the Western Alps 1964-1965. biotopes consisting of mountain torrents, often rich in small waterfalls and pot-holes, deriving from glaciers (No. 21 , 27, 44A, 47, 53, 71 , XXXV) or from lakes (No. 22); some are spring-brooks in pastures (44, 71 , XXXIII). Station No. 62 is a Iimnocrenic spring with an effluent brook, and 88 - , .... ~u " 'H' : 00 , ~...... ;; ! h"11 111 ] ~-~~ II .. 0 .In •• ' ..... 1 ~ :- ~ •• 1, .It •• ; , It~ ~ •

~ ~ ~ 11I' I J '1' "",----,,1 '_____"___=__' -Hl- C, ~':';";':.::.~I- + ~ ~~mmmTfr1Ji1i ~...f: ::t j L .... '~{..J:.::..!.....1 • ,.. n., L'.. .'-+-1. + .or•• '.,. • " - ~""' -! . 11 ~' "", ,' "'",.. II ~1-11 _ ,.-.'1~-""...... :-+--+,'-:-...;._c , r--.--:''''+ ~~ - '''. t ~

.~ I,..J' -~_ ...... ""'_" 11 ,)I ·.. !--- . , ...., ,• I~-•• ,I .. , . , , . . ...,.-, .. ::tt!.~ 11 J '-1' + 'j_ 1'1 1~.1"J.114.....,- -1 - '-D 1 ""'01 ' 1.' 0 . t L' -I __

----r-- '~'.'I~' ,. ''' '.Ii'' .. i . .. , --r--'"'I>," -':;,+". ,0 J.!!.'.!!:!!-t .....----

u_~ooo~:.:~~II-FJ "" l~ "U_ T'---r.··---- ::'-.....:__L ' 0 (J D c ___ _ I.: ",;1' C II a p ,'!!~." , .~ ~ _, ,' .

J--!!-JIIO~ I~ 11 '~t'l ~ ---"-"-'---.."'. ..". ,~ ....,. ... i .r " ... • • Il ,!! ,. ".' '. •• , '. ,1 ~'A-t- ...... , '" III S'14" --l-"~: : 1_ ...' ::__ .... • • •• ... JlA . n!rO ...... Il" -~,---l-,t' &4 !:,.•• b. -i' ~n , IG J,U ' . U IIJ + T---'--IJ ,I ~ ~_ nl '!I~I"' ~ ..U J 1.1 'I ., ~. ... t - .

11 '''JO~.''I'U +501 -~-rt--:=t~ ...... + ~.+ :: , ...... " .. D t!..t IS 111 t • i tl t f-'i ; 4 ""'" n 1 • " lIt -+ -. ... ~~IICII_'_'!['O!.iI-.:J...:a t 13I' · ..... +-- ~.. _ ~~ _ ,.- : I : I t"'~~'+-U.U ' Jt .. j n!ill _ , O_ ' ,: , J +-~ .. a 0 ___ nil. :01' ; t ' _~ ---.!.....,....' 4u nOCl , .. 11115 ,,.1 4 .2 •• 11 :::,~" t ' .. I .. -11.',\' -...... ~ ,,-trio1 110 ' . *r- ~ - ""... .. " W.1 ...... ,.. ~r --r -:::- t -1-' . .- ~ :t14:n~a:Jt~}..2... "-.!GI~ i t" : : ,. :- 1' - J . , ~l'.50 _.'!.-'3 ...... ' .' +- .....!!...~~~~ I I.. il ' I I I -+-

1I'~"'1 I ~ 12 204. I ... 'i ' IS II, I~ • .. . 1 I I..·· I.- .. ·"~I'-, ..51 ...... " t - ~ ::...... ~ f-I' • • l ' " ..... ' " . ~.. r '~ 13 1401 !',." . T~ -.....,... ----:.-r-~! I l" h . 1-i - I .. I.... S --;"':"_U' !, .' 1.", ln,r _ ~ ,,_ !~.II) m." , , 1'1 111; +_0' r ~'~~~~ 15 n'lOfl~.~_.U _U .j. U t U J.U4 ..' '' , ~ ! ~ _ I-- f-- r t o 00 UT lDWI ' IM U II ju 1.1 nh.) ,..... ~ 1 .. ~. ' j , \0 ~.I!!J.SGr ... ,; II " " l I lOt .. . , .:.z (10 It' . 0 In P , I It ,,~ ." I., · , +-... ., I • I "_11'~ .~ ~...-- ".. .. ~-' ,. -I T- r ~O" .. I I 1 I 1.... ·• I 1 1 I 1-- 1 I 1 1 I TI I • ~ Fig. 2. Synoptic table showing the environmental factors of the investigated biotopes, the taxa and the stages recorded. station No. 32 is a waterfall. In all these biotopes the pH is just about the neutrality, the hardness is generally low (by 6°Fr. in the springs, to 16°Fr. in the falls), the dissolved oxygen is generally near saturation (min. 87% in August in station 22) or clearly above saturation during June and August, in coincidence with the falls or with the maximum growth of plants and algae on the bottom (max. 157% in August in station 53). In lenitic ambients dominate Limnephilids, mainly Limnephilinae (No. 17, 18,20) and Halesinae (No. 28, 28B, 29) (Fig. 2). The lotic ambients are characterized by the predominance of Rhyacophilidae (No. 21, 22, 44, 62) or of Drusinae and Halesinae (No. 47, 53). In the lakes with effluents we find also Drusinae with Limnephilinae and Halesinae (No. 21, 22, 63). Since the sampling in every station was not effectuated regularly through the seasons, we cannot have exact news about the periods and the length of the emergence. Nevertheless we can say that, at this altitude, the emergence concentrates in summer and early autumn. Table 1 gives some clearer information on the distribution of genera (only adults). In running water: Rhyacophila, Apatania, only in station 21; Drusus in rapid water; Cryptothrix very abundant in station 27. In lenitic waters: Limnephilus, very abundant in station 20; Halesus and Allogamus in lakes with their tributaries and effluents, but especially in running waters of station 47. We can conclude that the orobiontic trichopterous fauna consists mainly of Limnephilidae and Rhyacophilidae. Therefore the populations of alpine running waters or small lakes of high altitude, are mono- or pauri-specific, but only the complete study of the gathered material will furnish a global table of alpine trichopterous fauna. In regard to single species we note: (1) The specimens of Limnephilus rhombic us of station No. 20 are bigger than those of the Central Apennines and, above all, the pigmentation of zones which delineate the 'fenestrate spot' of the anterior wings is extremely marked. Moreover also the sclerotized parts of the larvae have a particular pigmentation. (2) A. mendax and O. albicorne have been caught at high altitude. (3) R. glareosa, R. kelnerae and L. extricatus are here signalized for the first time in Italy, even if they were already known in the Alps. The identification of R. kelnerae leaves some doubt, since the shape of the superior lobe of the phallus does not correspond exactly to the drawing by SCHMID (1971). We will enter into a discussion upon this problem in a following paper. (4) We have to consider as orobiontic species: Apatania fimbriata, Drusus discolor, Monocentra lepidoptera, Halesus rub• ricollis, Allogamus mendax. (5) According to geographical distribution, in the gathered material we can distinguish: (a) Holarctic terms (L. rhombicus, L. stigma), (b) Eurosibirian terms (L. coenosus), (c) European terms (P. ludificatus, P. grandis, L. extricatus, O. albicorne), (d) Southwest European terms (D. discolor), (f) Pyrenaic• alpine and Central European mountain terms (A. fimbriata), (g) Alpine and Central European mountain terms (W. copiosa, H. rubricollis, A. mendax), (h) Alpine (and Pyrenaic?) terms (c. nebulicola), (i) Sardinian - Apuanian Alpine - Western Alpine terms (M. lepidoptera). 90 Table 1. Western Alps Trichoptera: list of taxa and number of adult specimens (00+ <;?<;?)found above 2000 m. s.1. (1964-65)

station nO 17 18 20 21 22 27 28 2~A 28B 29 44 44A 47 53 62 63 65 71 XXXII XXXIII XXXV len. len. len. lot. lot. lot. len. len. len. len. lot. lot. lot. lot. lot. len. len. lot. lot. lot. lot. TAXA lot. alt.m.s.l.m. 2300 2220 2000 2020 2000 2060 2500 2550 2500 2460 2000 2000 2000 2450 2041 2400 2000 2050 2050 2000 2700

Rhyacophila glareosa McL. 16 Rhyacophila intermedia MeL. 8 Rhyacophila kelnerae Seh. 2 4 2 Rhyacophila torrentium Pic!. Rhyacophila tristis Pict. 3 Rhyacophila vulgaris Pic!. 4 11 Rhyacophila sp. Philopotamus ludificatus McL. 2 9 2 Wormaldia copiosa MeL. Wormaldia occipitalis occipitalis Piet. Plectrocnemia sp. 'i' Polycentropus sp. Apatania jimbriata Pic!. 18 Drusus discolor Ramb. 2 6 4 2 2 Drusus sp. 1 Cryptothrix nebulicola MeL. 45 4 Monocentra lepidoptera Ramb. Limnephilus extricatus MeL. 2 Lemnephilus gr. hirsutus Pic!. 'i''i' Limnephilus rhombicus L. 12 Limnephilus stigma Curt. 26 Limnephilus sp. 'i''i' Halesus rubricollis Pic!. Halesus sp. 'i''i' Allogamus auricollis Pic!. Allogamus mendax MeL. 2 22 Allogamus sp. \0 Odontocerum albicorne Seop. 7 ...... Summary

This preliminary note about a wide trichopterological research program in the Western Alps shows the species found above 2000 m. The list is still approximate, which will be completed by forthcoming findings. In a series of 45 inspections made from August 1964 to August 1965 in 21 stations we have not only looked for a collection of species, but our attention was also directed to the ecological aspect of the research. In consequence we have determined in the single biotopes (Iotic and lenitic) environmental factors such as: temperature, pH, hardness, oxygen content. Up to date, 33 taxa of aquatic stages and adults have been identified. They include orobiontic and orophilous terms. A survey on their zoogeographical categories is given.

References

BERLAND, L. & MOSELY, M. 1936-1937. Catalogue des Trichopteres de France. Ann. Soc. Ent. France. 105-106: 101-144, 133-168. BOTOSANEANU, L.: Trichoptera. In J. ILUES.: Limnofauna Europaea. pp. 285-309. Stuttgart (Fischer) 1967. FISCHER, F. C. I. 1960-1973. Trichopterorum Catalogus. Amsterdam (Nederlandsche En• tomologische Vereeniging). McLACHLAN, R. 1874-1880; 1884. A monographic Revision and Synopsis of the Trichoptera of the European Fauna. First additional Supplement. London. J. Van Voorst. MORETTI, G. P. 1937. Tricotteri della Venezia Tridentina 1921-1935. Studi Trentini Sc. Nat. 18: 43-73. --. 1937. Una nuova larva di Drusus Steh. e la sua posizione sistematica (Trichoptera Limnophilinae). Boll. Soc. Entom. Ital. 68: 20-24. --. 1938. Tricotteri della Valsesia. Monografie del Comitato Scientifico. c.A.1. Varallo Sesia pp.49-66. MORETTI, G. P., CiANFICCONI, F., GIANOTTI, F. S., PIRISINU, Q. & VIGANO', A.: Informazioni sui Tricotteri delle Apuane. In BACCETTI, B. (Lavori della Societa Italiana di Biogeografia. N.S.l): II popolamento animale e vegetale delle Alpi Apuane. pp. 488-532. ForH 1970. SCHMID, F. 1971. Un nouveau Trichoptere des Alpes fran!;aises. L'Entomologiste. 27: 28-30.

Discussion HIGLER: Did you find Hydropsyche not at all? VIGANO: Not at all. These 21 stations are situated over 2000 metres. Perhaps in lower regions they can be found. ZINTL: Did I understand you properly, Potamophylax is in a region without trees and without big leaves? Do Potamophylax and Stenophylax also live in a region above the tree line? MORETTI: Qui, nous avons trouve les larves ici au fond des ruisseaux; les adultes ont ete trouves souvent plus en bas, en zone a Larix et latifoliae.

92 Proc. of the First Int. Symp. on Trichoptera,1974, Junk, The Hague

The taxonomical and ehorologieal problem of Drusus impro• visus MeL. in the North-Central Italian Apennines

GIAMPAOLO MORETTI & FERNANDA CIANFICCONI

McLACHLAN in 1884 established Monocentra improvisa from some specimens found in Central Italy (Appennino Pistoiese). In the study of SCHMID on the subfamily Drusinae (1956), M. improvisa is attributed to the mixtus group of the genus Drusus, as the affinity with the genus Monocentra consists only on the presence of scales in the androconial pocket of the 0 posterior wing. While examining the distribution of Drusus improvisus, we found this species from the Appennino Tosco-Emiliano to the Appennino Umbro-Marchigiano (besides the Apuanian Mountains). Furthermore, two other taxa similar to the above mentioned species have been singled out: one appears in the Appennino Umbro-Marchigiano, especially on the Adriatic slope (Mt. Catria, Mt. Cucco, Mt. Cavallo) and the other in the Appennino Abruzzese (Mt. Maiella, Mt. Como) (Fig. 1). We designate provisionally the first as taxon 1 and the second as taxon 2. Also in both these new taxa the spur-formula is 1,3,3 (0 and <;') and the 0 presents scales in the posterior wing pocket. However, the differences between the D. improvisus and the two above mentioned taxa are rather evident and consist in both morphology of the 0 genitalia, the androconial pocket of the posterior wing and variations in size, colour, time of emergence. D. improvisus is brownish-black in colour with dark palpi; the specimens of taxon 2 are dark brown with palpi of a lighter colour, they are also the largest in size; the specimens of the taxon 1 are the most paler in colour and the smallest. The emergence of D. improvisus generally takes place in summer, while the taxon 1 emerges in winter and the taxon 2 in autumn.

Genitalia 0 of Drusus improvisus The 8th tergite shows a large elongated triangular area covered with short stout spines (Fig. lA). The apical edge forms 2 obtuse lobes (which are not always visible); the short stout spines are large and spaced (Fig. 2C). The underlying membranous areas are large, white and triangular and are shaped like an overturned sail (Fig. 2A). The superior appendages are rather large, protruding and oval-shaped, if viewed from the side (Fig. 2A), if observed from inside, they appear rounded and concave (Fig. 2B,C).

93 M

FORLI ::>1'< · ...iC L o o '0" It o <'0 UCCA

R~N/r . ARE ' .,

It

l'!I U I

,~

VITE~BU ....i.l 0.,

~, ~(.~ , ...... ~ . .'...... " R(rv'A

Fig. 1. Drusus improvisus MeL. and related taxa: distribution map in North-Central Italy. c Fig. 2. Drusus improvisus McL., male genitalia: (A) lateral, (B) caudal, (C) dorsal (after SCHMID). 95 The intermediate appendages are narrow with apex upward hooked (Fig. 2A). The inferior appendages are large, prominent, conical and slightly concave in the inner part (Fig. 2A,B,C).

Genitalia 0 of Drusus taxon 1. The apical edge of the 8th segment shows 3 distinct lobes: compared to the two lateral lobes, the middle one is elevated and situated backward (Fig. 3A,C). The short stout spines are compact on the three lobes, but spaced out around them (Fig. 3C).

3 B

c Fig. 3. Drusus taxon 1, male genitalia. Explanation as in Fig. 2. 96 The pale underlying membranous areas have always been observed as narrower and more sinuous than those of D. improvisus (Fig. 3A). The superior appendages are large and oval, from a side view they sometimes appear hatchet-shaped (Fig. 3A), if seen from the inner side they are concave (Fig. 3B,C). The intermediate appendages are wide, stumpy and do not show an upwards hooked apex (Fig. 3A), if seen from above they are short, rounded and divergent (Fig. 3C). The inferior appendages are large, subcylindrical and rounded at the apex (Fig. 3A,B,C). Genitalia 0 of Drusus taxon 2 The apical edge of the 8th segment shows 3 almost similar lobes, compared to the two lateral lobes, the middle one is situated only slightly backwards (Fig. 4C). If seen from the front, these lobes are less evident than those of the taxon 1 (Fig. 4B), also the short stout spines are less extended backwards (Fig. 4C). The pale areas are small and irregular (Fig. 4A). The superior appendages are large and oval-shaped. The intermediate appendages are short, strong and, if seen from the side, show a rounded apex (Fig. 4A), if observed from above or from the front, they are widely divergent (Fig. 4B,C). The inferior appendages are large, conical and angulate towards the middle of the upper edge (Fig. 4A,B); seen from below, they are longer and more cylindrical than in the case of D. improvisus (Fig. 4C). o Posterior wings The characters observed on the 0 posterior wings are summarized in the table and include the data of 100 specimens for each taxon. Drusus Drusus Drusus improvisus taxon 1 taxon 2 Posterior wing length mm 7.6 6.9 8.5 Androconial pocket length mm 2.4 1.4 2.2 Scale number 170 52 58 Scale length fL 130 126 156 Scale row number 3-2 2 2 Periandroconial nervure terminal position Parallel Divergent Divergent (Fig. 5) (Fig. 6) (Fig. 7) Distribution of scales In light area In dark area In dark area surrounded by surrounded by surrounded by dark area light area light area (Fig. 8) (Fig. 9) Ecology and behaviour The larvae and pupae of all three taxa are cn~nobiontic and rheophilic. They live in small springs of the Central Apennines, especially at heights of between 400 m and 1300 m, where the annual mean water temperature is between 6 and 10°C.

97 4 B A

c Fig. 4. Drusus taxon 2, male genitalia. Explanation as in Fig. 2.

The mature larvae, interstitial during daytime (up to 20 cm in depth) become epigeal at night and go upstreams. The young larvae are also hygropetric (e.g. D. improvisus of the Tiber spring). The pupae, enclosed in cylindrical sand cases, are hidden in the bottom. The adults never leave the area surrounding the springs and sometime penetrate into water intake tunnels. The gelatinous spawn masses, containing about 100-150 eggs each, are laid on rocks and stones of the springs; the Drusus of the taxon 1 spawns even with snow in winter.

98 99 100 101 102 103 On the basis of these considerations about the morphological differences found in "" and considering also the different times of emergence and of spawning as well as the geographic distribution, the taxonomic problem of Drusus improvisus and of the satellite taxa may be set out in the following terms: (a) Taxon 1 and taxon 2 could be two new species; (b) Taxon 1 and taxon 2 could be two new subspecies of the Drusus improvisus; (c) It could be that one of the two new taxa is a subspecies of the other. Work is still being carried out in search for other distinctive characteristics (especially on 'il'il, on larvae and on pupae), but for this Symposium, witnessing the presence of so many famous Trichopterologists, we have thought it convenient to present our observations beforehand, hoping to receive helpful suggestions. 1

References

BOTOSANEANU, L. 1960. Trichopteres de Jugoslavie recueillis en 1955 par Ie Dr. SCHMID. Deutsche Ent. Z. 7: 262-293. KUMANSKI, K. 1973. Die Unterfamilie Drusinae (Trichoptera) in Bulgarien. Tijdsch. Entom. 116: 107-121. MARINKOVIC-GOSPODNETIC, M. 1971. New species of Trichoptera from Bosnia and Her• cegovina. Bull. Scien. A. 16: 144. McLACHLAN, R. 1874-1880. A monographic revision and synopsis of the Trichoptera of the European fauna. London I. V. Voorst. (1884) First additional supplement MORETTI, G. P. & CIANFICCONI, F. 1964. Sulle formazioni androconiali di alcune specie di Tricotteri. Atti Acc. Naz. It. Entom. Rend. 11: 199-202. --. 1965. Androconi presso ia subfamiglia Drusinae (Trichoptera). Boll. Zool. 32: 975-986. --. 1970. Ciclo bioiogico e comportamento etologico di un tricottero fonticolo dell'Appen- nino Marchigiano (Drusus sp.n.?). Boll. Zool. 37: 505. SCHMID, F. 1956. La sous-famille des Drusinae. Mem. Inst. Roy. Sc. Nat. Belgique. 55: 1-92.

Discussion Ross: It seems to me that whenever we start looking at local populations very closely we begin tum up many more new species. FLINT: I vote for three species. But the Europeans must decide what names are to be applied to them.

1 These instances are considered by Ross (i.l.) as perfect examples of Pleistocene speciation.

104 Proc. of the First Int. Symp. on Trichoptera, 1974. Junk, The Hague

Ecologie et reproduction chez les Trichopteres cavernicoles du groupe de Stenophylax (Limnephilidae, Stenophylacini)

YVEITE BOUVET

Rappel systematique (SCHMID, 1955): Limnephilidae, Limnephilinae, Stenophyla• cini, groupe de Stenophylax, genres Stenophylax, Micropterna et Mesophylax. Dans ces trois genres, seulement treize especes ont ete observees regulierement dans Ie milieu souterrain: Stenophylax crossotus, S, mitis, S. mucronatus, S. permistus, S. vibex vibex, S. vibex speluncarum, Micropterna fissa, M. lateralis, M. nycterobia, M. sequax, M. testacea, Mesophylax aspersus, M. impunctatus. Definition du milieu souterrain: L'etude des Trichopteres du groupe de Stenophylax conduit it donner une definition plus ecologique que spatiale du milieu souterrain. II comprend tout d'abord les cavites et fissures naturelles des massifs calcaires, mais aussi des roches granitiques et volcaniques. Nous pouvons egalement inclure dans ce do maine les cavites artificielles (tunnels, galeries, souterrains ...), certains eboulis de montagne, les chaos morainiques ou tectoniques et tous les milieux dont certaines caracteristiques climatiques (temperature, humidite, stabilite de l'air) favorables it l'estivation des adultes de Trichopteres du groupe de Stenophylax miment Ie biotope cavernicole.

Cycle biologique (Fig. 1) II se caracterise par une phase aerienne cavernicole, au cours de laquelle les adultes estivent dans des milieux souterrains o~ proches du milieu cavernicole, dont la temperature, sous les latitudes temperees, se situe entre 0 et 14°C et la teneur en vapeur d'eau de l'atmosphere voisine de la saturation (90 it 100%). La phase larvaire hivernale se deroule dans les ruisseaux epiges, temporaires Ie plus souvent. II existe une seule generation par an. La ponte, deposee it l'automne, sous des pierres emergees, donne naissance it des larvules de 1e stade en Octobre• Novembre. Les cinq stades larvaires et Ie stade nymphal se deroulent en 7 mois environ. A I'emergence (Mai), les adultes migrent, parfois sur de longues distances (10 it 20 km selon BITSCH & FROCHOT 1962), vers les lieux d'estivation. Les femelles it l'emergence ont des ovaires peu developpes. II faut attendre Ie mois de Septembre pour observer des oeufs prets it etre pondus. Nous avons ici un phenomene de reproduction differee, qui s'apparente it une diapause imaginale ovarienne.

105 ,.• .". ,- .". / stade larvaire / / 2 0 stade 1. Pha8e aerienne I I aaverniaoZe I I 3 0 stade 1. I I Phase aquatique •I , epigee

stade 1.

stade nymphal Fig. 1.

Reproduction differee Les ovocytes, a l'emergence des femelles, sont au stade A (NovAK & SEHNAL, 1963) et restent ii ce stade pendant 6 a 8 semaines. La vitellogenese intervient alors et, apres deux semaines, les oeufs arrives a maturation descendent dans l'oviducte (BOUVET, 1971). Deux exceptions sont signalees: SVENSSON (1972) observe chez les femelles de M. lateralis et de M. sequax a l'emergence des ovaires au stade B ou C. DENIS (communication orale) obtient en elevage des femelles de M. sequax dont le~ ovaires sont bien au stade A, mais il observe un developpement ovarien apres trois semaines et la ponte un mois apres l'emergence. Dans Ie premier cas, la latitude tres septentrionale a laquelle ont ete faites les observations peut influencer Ie cycle des especes considerees. Dans Ie second cas (DENIS), nous ne pouvons encore interpreter ce resultat. Influence des conditions de milieu 1. Phase imaginale du cycle Le sejour dans Ie milieu cavernicole des imagos du groupe de Stenophylax leur impose des facteurs du milieu strictement definis en fonction de la latitude et de l'altitude. Nous avons choisi d'etudier, dans un premier temps, la temperature et la lumiere. a. La temperature: Les Trichopteres, du groupe de Stenophylax colonisent des milieux dont la temperature, dans les zones temperees, ne depasse pas 12°C, sauf pour M. aspers us que I'on recontre souvent jusqu'a 14°C. Les variations thermiques journalieres dans Ie domaine souterrain sont tres faibles ou nulles. La limite 106 inferieure des temperatures supportees par les adultes est tres basse, surtout dans les grottes d'altitude et peut atteindre O°c. De 0 it 14°C, nous observons un parfait synchronisme dans la maturation ovarienne; la vitellogenese se declenche, pour les memes especes, simultanement dans les grottes it O°C et dans celles it 14°C (BOUVET, 1971). Ces resultats sont confirmes experimentalement. Au laboratoire, une temperature superieure it 14°C appliquee it des elevages d'imagos n'accelere pas la maturation ovarienne. Par contre, Ie taux de mortalite s'eleve et l'oviposition est· anormale. Les oeufs obtenus dans ces conditions ne se developpent pas. A la suite d'observations effectuees au Tassili n' Ajjer (Sahara sud-oriental) ou I'on rencontre une importante population de M. impunctatus (MICHALON, 1973) qui supporte une temperature souvent superieure it 20°C, nous avons etudie Ie preferen• dum thermique et la temperature lethale superieure des especes cavemicoles qui peuplent la zone temperee. Preferendum thermique: sur un gradient thermique de 0 it 25°C, les imagos du groupe de Stenophylax choisissent les temperatures hautes, 19 it 20°C et 23 it 24°C (BOUVET, 1975). Temperature lethale superieure: une temperature de 35°C, appliquee aux imagos pendant 24 heures entraine une mortalite de 50% dans une atmosphere saturee en vapeur d'eau. b. La lumiere: dans une grotte, les Trichopteres ne semblent pas choisir leurs lieux d'estivation en fonction de l'intensite IUlnineuse, mais en fonction de la temperature et de la teneur en vapeur d'eau de I'atmosphere. lis peuvent sejoumer it I'obscurite complete, mais aussi dans des zones ou penetre une lumiere diffuse. Au laboratoire, des elevages pratiques it I'obscurite complete ainsi qu'en photoperiode naturelle donnent les memes resultats. Les femelles deviennent capables de pondre apres la meme periode de diapause ovarienne. La photoperiode ne parait pas etre Ie facteur determinant dans la levee de la diapause. En conclusion, la diapause ovarienne doit etre programmee pendant Ie developpement larvaire.

2. Phase larvaire du .cycle Les larves peuplent, dans la plupart des cas, des ruisseaux temporaires peu profonds, ou Ie courant est rapide en periode de crue. Les eaux renferment entre 10 et 11 mg/1. d'oxygene dissous ce qui, aux temperatures qui regent dans ces ruisseaux (8 it 12°q correspond it la saturation. La durete totale est de 250 ± 50 mg/1. C03Ca et Ie Tac compris entre 200 et 250 mg/1. C03Ca. Ces cours d'eau s'assechent souvent avant l'emergence. Dans ces conditions, les larves s'enfoncent dans Ie sous-ecoulement lorsqu'elles Ie peuvent. One temperature basse (3 0q, appliquee it partir de l'oeuf, ralentit beaucoup Ie developpement larvaire; Ie deuxieme stade larvaire est atteint apres une annee. A 15°C, il n'y a pas d'acceleration de la vitesse du developpement larvaire, mais une accentuation des variations individuelles. De ce fait, les emergences s'etalent sur une plus longue periode (BOUVET, 1975). A cette temperature, les femelles it I'emergence presentent

107 des ovaires au stade A. La temperature lethale superieure (50% de mortalite en 24 heures) est comprise entre 28 et 30°C suivant les especes. La temperature it laquelle sont soumises les larves ne parait pas etre Ie facteur programmant la diapause ovarienne. La photoperiode appliquee durant Ie developpement larvaire pourrait reguler Ie cycle reproducteur et programmer la diapause. Mais cette hypothese est en cours de verification.

Discussion

MALlCKY: What happens with the males, as concerns the diapause? BOUVET: Les males ne semblent pas concernes par la diapause, j'ai realise des dissections d'appareils genitaux des males. lis semblent etre capables des l'emergence de fournir aux femelles, les elements necessaires it la reproduction; je ne sais pas si M. DENIS a des elements plus precis 11 ce sujet. DENIS: I have thought to this problem but I have not yet studied it. BOUVET: Mais il semble en fait que les males ne sont pas en cause dans I'impossibilite de ces especes 11 se reproduire des I'emergence. MALICKY: I think it should not be too complicated to have a look at the spermatogenesis. This is, as far as I know, totally unknown. Nobody has investigated it in these species. Another remark. You said that the adults are in total darkness in the cave. I made investigations in a cave in Lower Austria in which a population of about 3000 to 5000 specimens of Microptema nycterobia is found every summer. it was my impression too that they are in total darkness. The human eye cannot see anything, and I could not find any trace of light even by using instruments. But I exposed photographic material for one day in June. The film exposed in the region where the caddis-flies were abundant was blackish after processing. The film exposed deeper in the cave showed no trace of blackish, and there were no Trichoptera too. From this it was clear that the adults sit only in the region of the cave which is reached by very slight traces of light. So I cannot exclude the possibility that day length may influence the development of the ovaries, but the sensitivity level of these species must be very different from others, e.g. Limnephilus etc. RIGLER: Est-ce qu'il y a connu des cas dans les quels sont des larves et des nymphes dans les eaux dans les grottes. BOUVET: Dans les grottes il a ete recolte des larves et des nymphes, en particulier en Belgique; mais il s'agit toujours de ruisseaux qui entrent dans une grotte et qui entrainent par derive, les larves et les nymphes dans les grottes, mais generalement une population ne peut pas s'installer dans la grotte aux stades larvaires et ces larves et ces nymphes ne peuvent pas achever leur deve10ppement dans la grotte. MORETTI: Vous etes en train de I'etudier experimentalement? Vous pouvez avoir une dis• tribution des adultes dans la region qui est comprise entre 14 et 20°C pendant que dans la nature on a generalement peu d'individus dans cette region. Vous avez mesure I'humidite relative de cette zone? BOUVET: Pour vous preciser les conditions d'experience: c'etait une gouttiere, et au fond de cette gouttiere j'avais essaye de reproduire les conditions des parois des grottes: j'avais mis des pierres, de I'argile, du gravier. Et je maintenais en permanence cette couche completement humide et Ie meme taux d'humidite que j'ai realise avec les moyens dont je dispose maintenait toujours l'atmosphere 11 saturation. II y avait toujours sur les parois des gouttelettes de condensation.

108 DENIS: Tout d'abord pour repondre a M. MALICKY, je dirai que la spermatogenese etudiee notamment par M. LE LANNIC, se situe au cours du 5e stade et se termine au debut de la vie nymphale. Vous dites, Madame BOUVET, que I'oviposition peut intervenir 6 a 10 semaines apres l'emergence. Pour moi, un delai de 6 semaines signifie qu'il n'y a pas eu de diapause. En efIet les femelles etant initialement immatures, il faut un certain temps pour que la maturation se produise. Et j'ai obtenu des generations sans diapause pour Limnephilus lunatus, L. rhombicus, L. auricula, L. bipunctatus et L. vittatus. Or dans ces cas, l'oviposition n'intervient que un mois et demi apres l'emergence. Par contre, lorsq'il y a diapause, Ie delai est nettement plus long: 3 a 4 mois. BOUVET: Je pense que vous faites une erreur parce que vous considerez que Ie fait d'avoir un arret de deve\oppement qui ne dure que 6 semaines constitute une preuve de non-existence de la diapause, car une diapause est caracterisee par un stade d'arret a une phase bien precise, sans que la duree de cet arret intervienne. Or, la valeur totale du temps entre l'emergence et I'oviposition n'est pas du tout un argument de l'absence de diapause, car ici vous avez un arret tres long, a un stade bien determine et puis tres rapid~ment, vous avez un bouleversement total avec declenchement de la vitellogenese, et tous les phenomenes de la vitellogenese intervien• nent pratiquement en 8 jours, en une semaine, alors que pendant les 10 semaines precedentes, je crois pouvoir vous dire qu'au niveau du corps adipeux, rien ne se passe, ni au niveau de I'ovocyte. DENIS: J'ai commence a determiner la chronologie du developpement ovarien de M. sequax. Chez des imagos maintenus a la lumiere du jour et a 18°C, la vitellogenese a commence environ 3 semaines apres l'emergence (etude faite a l'aide de dissections) et les pontes sont intervenues 3 a 4 semaines plus tard. Dans ce cas, a mon avis, il n'y a pas eu de diapause. Par contre d'autres imagos ont ete eleves a I'obscurite complete et a 8°.C. Actuellement, deux mois et demi plus tard, la viteUogenese n'a pas encore commence. Pour moi, dans ce dernier cas, il y a diapause. Je voulais dire aussi que Ie probleme de la rupture de la diapause semble particulier chez les Trichopteres cavernicoles. Pour 1es especes etudiees par NovAK: Limnephilus, Glyphotaelius, Anabolia, c'est une longue photoperiode qui declenche la diapause et une diminution de la photoperiode qui rompt la diapause. Mais par ailleurs j'ai constate chez L. rhombicus, Halesus radiatus et Anabolia nervosa que des animaux mainte'nus longtemps dans les conditions de la diapause, rompaient malgre tout leur diapause. Cette rupture intervient apres 3 a 4 mois pour les adultes de L. rhombicus, 4 a 5 mois pour les larves de H. radiatus et ' environ six mois pour celles de Anabolia nervosa. Et je pense que c'est Ie meme phenomene qui se produit, mais de fa90n naturelle pour les Trichopteres cavernicoles. Selon moi, chez ces derniers, la rupture de la diapause interviendrait sans modification des facteurs externes declenchants. BOUVET: Bien sur, parce que je pense que ce qui declenche la diapause s'exerce sur Ie stade larvaire donc sous terre, durant Ie stade imaginal, Ie milieu n'est plus Ie meme, il n'intervient donc pas de modification de ce facteur. De toute fa90n, je pense que la levee de la diapause intervient par une sommation du temps a partir du moment du declenchement de cette diapause il y a sommation au bout d'un temps determine - car queUe que soit la latitude ou I'on se place, des pays temperes jusqu'aux zones sahariennes, Ie cycle se deroule toujours dans Ie meme temps sur 12 mois. Les differentes phases du cycle sont synchrones. En consequence, je pense que c'est tres certainement une sommation qui intervient plus qu'autres choses. Parce que les conditions auxqueUes sont soumises les individus, par exemple en climat desertique et en climat tempere, sont tres differentes. Je pense que c'est dans cette direction que se trouve I'explication de la levee de la diapause, si diapause il y a, les specialistes ne sont pas d'accord sur les definitions du terme!

109 Proc. of the First Int. Symp. on Trichoptera, 1974, Junk, The Hague

The Trichoptera population of a temporary ecosystem of the Umbrian Apennines (Perugia, Italy)

G. P. MORETII, F. CIANFICCONI & Q. PIRISINU

In the Central Apennines of the Italian Peninsula the Sibillini Mountains constitute an important massif. The highest peak is Mount Vettore (2478 m above sea level). At the height of 1300 m there is a vast karst plain, the 'Piano Grande', which is about 7 km long and 3 km wide. It is crossed by the 'Mergani' ditch collecting astatic waters which is surrounded by numerous do lines of various sizes. In summer these are completely dried out, while in winter they are frozen. Thus the water-system depends on the spring and autumn rains and on the snowfall in winter. The karst ecosystem shows both lotic and lentic environments; in summer small pools are left behind in the Mergani ditch which, at the end of its course, disappears into a sink-hole. From an extensive limnologic work on this ecosystem, we have chosen the part represented by caddis-fly fauna, on which we would like to propose someinforma• tion. Our research is still going on, therefore we can only report what we observed from November 1973 to August 1974. In the Piano Grande we can distinguish two main types of aquatic associations: the lacustrian type (Potametea) which charac• terizes the Mergani water-pools (Potamogeton natans and Ranunculus trichophyllus), and the marshy type (Magnocaricion) which qualifies the do lines and ditch (Caricetum) . Minimum and maximum chemio-physical values (1973-1974) are enumerated below: Min. Max.

T. air °C -1 28 T. water °C 0 24

O 2 s.v. 40 130 Total hardness (OF) 4 18 pH 6 8.5 KMnO. cons. mg/1. 3.5 16.5 N03 - mg/1. 0.44 5.28 N02 - mg/1. 0 0.066 SO;- mg/1. 2 70 PO~- mg/1. 0.05 6 Cl- mg/1. 12 110 Si02 mg/1. 0 5 Fe+ mg/1. 0 4 Mn+ mg/1. 0 1.7

111 According to these results one notes considerable oscillations of temperature and of the oxygen dissolved in the water, as well as great variation of pH, total hardness, organic substances, nitrates, sulphates, phosphates and chlorides; therefore, in the flood period, it is a typical environment for the eurybiont caddis-flies. Thus, the species we have found there are exactly those we expected to find in such a biotope and precisely as follows: (1) Agrypnia varia FBR.; (2) Limnephilus bipunctatus CURT.; (3) Limnephilus flavicornis FBR.; (4) Limnephilus sparsus CURT.; (5) Limnephilus vittatus FBR.; (6) Grammotaulius atomarius FBR.; (7) Micropterna nycterobia MeL.;' (8) Leptocerus tineiformis CURT. On the Mergani hydrographic map, one can see the zonal distribution of the species, but above all one can see that Gr. atomarius is the most frequent and diffused term in the biotope (up to 360 specimens in 1 m 2). The larva of this limnephilide builds its cases almost exclusively with Carex buxbaumi. Limnephilus flavicornis follows it with a much lower population density (24 specimens in 1 m 2). The other taxa are much less numerous. Micropterna nycterobia, which always lives at a certain height in the Apennines, dwells only in the running waters of the sink-hole. Species that emerge first (April-May) are M. nycterobia and Gr. atomarius; the latter, as an adult, is still frequent in the meadows in August (00, SlSl). A. varia and L. flavicornis are the last ones to emerge. What concerns the diapause of some species, in these temporary waters, the situation is still to be cleared; however, we observed that, at the end of July (1974), about 70% of the females show an advanced vitellogenesis. Moreover, we could establish that A. varia lays her eggs in August, with a quick hatching of the larvae. These observations seem to agree, at least in part, with NovAK & SEHNAL (1965) when they affirm that one cannot observe diapause in the females of species having diapause and living at a high altitude, since their gonads are ripe at the beginning of summer. We must point out that three species A. varia, Gr. atomarius, L. flavicornis are characteristic of marshes, ponds and temporary pools of the Apennines plains at the height of about 1000 m, even though these species can be found frequently at much lower altitudes. In the intestine of larvae there has been found Gregarinida: in that of A. varia, Asterophora heeri; in that of L. flavicornis, Gregarina limnophili and in that of Gr. atomarius, Gregarina limnophili and Pileocephalus glyphotaelii.

References

AISA, E. & TATICCHI, I. M. 1964. Fisionomia delle zoocenosi rivierasche del Trasimeno nel corso dell'aumento dellivello. Boll. Zool. 31: 1403-;-1420. BRAY, R. P. 1971. Factors affecting the distribution of some Phryganeaeid (Trichoptera) in Malham Tarn, Yorkshire. Freshw. BioI. 1: 149-158.

, In winter 1974-1975, in the sink-hole young larvae of Microptema fissa McL. also has been found; so there are 9 different species of caddis-flies in the Mergani.

112 M. SIBILLINI- PIANO GRANDE MERGANI'S SYSTEM m.1300 (Umbria, Italy) Ji. Lotic. N. Lentic. ga dolines '" specimens/m' / SPECIES 1 • AgrypnlO vana Fbr 2 . Llmnephllus blPunctatus Curl 3 • flovlcornls Fbr 4 $ sporsus Curt

5 .to vlttatus Fbr : ..... 6.,. Grammataultus atomorius Fbr ' .. ' 7 • Mlcropterna rrycterobtl Me L 8 = Leptocerus tinelformls Curt

~)' , ) ( ~

Fig. 1. 113 CHAMPEAU, A. 1966. Contribution a I'etude ecologique de la faune des eaux temporaires de la haute Camargue. Arch. Ocean. Limno!. 14. DI GIOVANNI, M. V. 1964. Stabilita e modificazioni delle comunita epifitiche del L. Trasimeno durante I'innalzamento dellivello. Boll. Zoo!. 31: 1372-1385. FISCHER, F. C. 1. 1943. De Nederlandsche sort en von het genus Limnephilus (Trichoptera). Entom. Bericht. 11: 96-100. HICKIN, N. E. 1943. Larvae of the British Trichoptera, Limnephilus flavicomis L. Fabr. Roy. Entom. Soc. London. A. 18: 6-10. --. 1943. Larvae of the British Trichoptera, Limnephilus vittatus. Fabr. Proc. Roy. Entom. 'Soc. London A. 18: 72-74. --. 1953. Larvae of the British Trichoptera, Phryganea varia. Proc. Roy. Entom. Soc. London A. 28: 39-40. --. 1954. Larvae of the British Trichoptera, Grammotaulius atomarius. Fabr. Proc. Roy Entom. Soc. London A. 29: 89-92. MALICKY, H. 1971. Trichopteren aus Italien. Entom. Zeitschr. 81: 258-265. MARCUZZI, G. & LORENZONI, A. M. 1971. Osservazioni ecologiche faunistiche sui popolamento animale di alcune acque carsiche dei dintorni di Trieste. Vie et Milieu 21: 1-58 (1970); 22: 1-32. MARLIER, G. 1947. Notes sur les Trichopteres, I. Les femelles du genre Limnephilus Leach. Bull. Mus. Roy. Hist. Nat. Belgique 23: 1-13. MORETTI, G. P., CIANFICCONI, F. & PIRISINU, Q. 1968. Struttura della faunula di Tricotteri del bacino di Monterosi. Boo!. Zoo!. 35: 362. --& DI GIOVANNI, M. V. 1970. Insediamento di Tricotteri nei laghi agricoli. Boll. Zoo!. 37: 506-507. --& TATICCHI, I. M. (1970). Come si presenta ora la popolazione tricotterologica dei litorali del L. Trasirneno. Boll. Zoo!. 37: 508. --, PIRISINU, Q., RAVIZZA, C. & FIORELLI, M. A. 1972. Tricotteri e coleotteri idroadefagi del L. di Ventina (Lazio-Rieti). Riv. di Idrobiologia 11: 79-100. NovAK, K. & SEHNAL, F. 1963. The development cycle of some species of the genus Lim• nephilus (Trichoptera). Act. Soc. Entom. Cechoslov. 60: 68-80. --, --. 1965. Imaginaldiapause bei den in periodischen Gewassern lebenden Trichopteren. Proc. XII Intern. Congr. Entom. London 434. PEDROTTI, F. 1959. Entomofauna acquatica della palude carsica di Pietra Rossa (Monfalcone). Atti 1st. Veneto Scienze Lett. Art. 117: 330. SARETCHNAJA, S. N. 1961. The larvae of the caddis fly Limnephilus sparsus Ramb. (Trichopt• era). 1st. Bio!. Serbatoi SSSR; 24-27. SCHACHTER, D. & CONAT, M. 1952. Note sur la faune des eaux temporaires de la Petite Camargue. Bull. Mus. Hist. Nat. Marseille 12: 7-13. SCHMID, F. 1957. Les genres Stenophylax KoL., Microptema ST. et Mesophylax McL. Trab. Mus. Zoo!. Barcelona 2: 1-51. TATICCHI, M. I. 1968. Vicende stagionali delle cornu nita litoranee del L. Trasirneno (1963-65). Riv. Idrobio!. 7: 195-302. WICHARD, W. & REICHEL, H. 1970. Zur Trichopterenfauna periodischen Gewasser. Nachr. Bayerisch. Entom. 18: 57-58. --, --. 1970. Zur Trichopterenfauna von Baggerseen. Nachr. Bayerisch. Entom. 18: 66-67. WIGGINS, G. B. 1973. A contribution to the biology of Caddis-flies (Trichoptera) in temporary pools. Life Scienc. Contrib. R. Ontario Museum 88: 1-28. WINKLER, D. 1961. Die mitteleuropaischen Arten der Gattung Limnephilus LEACH. ·Deutsch. Entom. Z.N.F. 8: 165-214.

114 Discussion

ZINTL: Have you observed flavicornis cases only without sand grains? MORETTI: Oui. Nous avons aussi trouve L. fiavicornis avec des petites coquilles, mais pas avec les grains de sable. Surtout avec des vegetaux. NIELSEN: The case of L. fiavicornis is extremely variable. It may sometimes be built of sand grains, but it is generally built with vegetables. HILEY: Comment avez vous fait les preparations pour les photographies des fourreaux? lis apparaissent plus droit que mes specimens de L. sparsus et L. vittatus. MORETTI: Verticalement, sur un fond d'etoffe en etat sec, parce que la lurniere, avec l'eau, cause des reflexions.

115 Proc. of the First Int. Symp. on Trichoptera, 1974, Junk, The Hague

The Trichoptera of the stony shore of a lake with particular reference to Tinodes waeneri (L) (Psychomyiidae)

N. V. JONES

This paper reports on the emergence of adult caddis-flies from the stony shore of Llyn Hendref, Anglesey (National Grid Reference: SH 398765) during the summers of 1963, 1964 and 1965. The larvae and pupae of Tinodes waeneri were collected from the same site and data from these collections are related to the information on the emergence of adults of this species. Llyn Hendref is an eutrophic lake lying in an alluvial depression at 61 m above sea level. It had an area of about 170 000 m-2 and some chemical data are given by REYNOLDSON (1958). The configuration of the lake is now altered as a result of increased drainage of the surrounding area.

Methods

Emerging caddis-flies were trapped in a floating box trap anchored over about 40 cm of water on the north shore of the lake. The trap was made of Perspex and measured 75 x 75 x 45 cm. Nylon net panels on two sides helped to reduce condensation on the inside. The bottom of the trap could be closed with a sliding panel to allow it to be taken ashore where the catch was removed with an aspirator. The trap was in operation during the following periods:

15 July to 13 November 1963 1 April to 26 October 1964 1 May to 14 September 1965

It was emptied about once a week in 1963, every two to three days in 1964 and every four to five days in 1965. About 100 larvae of T. waeneri were collected from the same site on each of 31 months. Stones 8-16 cm diameter were collected from water of 35-50 cm deep and carefully conveyed to the laboratory where the larvae (and pupae when present) were removed. The larvae were measured following anaesthesia in water containing menthol crystals. 117 ...... 00

Table 1. The emergence trap collections of the ten most numerous species of caddis flies to emerge from Llyn Hendref, Anglesey. The collections are grouped into lO-day catches.

MAY JUNE JULY AUGUST SEPTEMBER OcrOBER TOTAL 21-1 1-10 10-20 20-30 30-9 9-19 19-29 29-9 9-19 19-29 29-8 8-18 18-28 28-7 7-17 17-27 27-7 7-17 17-27

Tinades 1963 18.0 23.0 23.0 35.0 100 84.0 21.0 9.0 2.0 o 315 waeneri 1964 0 0 6.0 50.0 39.0 23.0 59.0 70.0 55.0 35.0 21.0 33.0 41.0 26.0 9.0 4.0 2.0 0 o 473 1965 0 7.0 20.0 29.0 30.0 35.0 36.0 41.0 25.0 13.0 18.0 16.0 6.0 3.0 279 Ecnomus !~~~}No specimens trapped or seen tenellus 1965 o 000 o 0 1.0 5.0 7.0 4.0 o o o o 0 17 Polycentropus 1964 o 8.0 6.0 0 1.0 5.0 7.0 5.0 0 2.0 2.0 2.0 3.0 7.0 1.0 1.0 o o o 50 flavomaculatus 1965 3.0 0 0 1.0 0 1.0 0 0 2.0 3.0 1.0 0 o 0 11 Cymus 1964 o o 7.0 69.0 13.0 56.0 76.0 86.0 91.0 58.0 9.0 31.0 36.0 7.0 2.0 0 o o o 541 trimaculatus 1965 o 2.0 13.0 16.0 23.0 17.0 19.0 36.0 12.0 8.0 17.0 15.0 5.0 3.0 186 Cymus 1964 o 11.0 13.0 1.0 0 0 0 0 3.0 3.0 000 o o o o o o 31 flavidus 1965 5.0 1.0 1.0 0 0 0 0 4.0 4.0 000 o 0 15 Limnephilus 1963 1.0 9.0 2.0 0 0 o o o o o o 12 marmoratus 1964 o o o 2.0 7.0 9.0 5.0 5.0 10.0 4.0 9.0 0 0 o o o o o o 51 1965 o o o 6.0 3.0 1.0 3.0 1.0 0 000 o o 14 Anabolia 1963 o 0 000 3.0 9.0 4.0 7.0 1.0 1.0 25 nervosa 1964 o o o o o o o o o o o 0 1.0 5.0 40.0 20.0 9.0 3.0 o 78 1965 o o o o o o o o o o 0 3.0 4.0 2.0 9 Mystacides 1963 o 0 o o o o 4.0 1.0 o o o 5 azurea 1964 o o o o 1.0 2.0 4.0 1.0 1.0 0 o 1.0 3.0 4.0 1.0 o o o o 18 1965 o o 2.0 3.0 o 0 2.0 0 0 o 2.0 9.0 3.0 1.0 22 Oecetis 1964 o o o o o o 0 1.0 2.0 2.0 o o o o o o o o o 5 lacustris 1965 o o o o 2.0 2.0 4.0 13.0 13.0 o o 2.0 2.0 o 38 Athripsades 1964 o o o 2.0 1.0 2.0 0 1.0'" 0 0 o o o o o o o o o 6 aterrimus 1965 o o o 1.0 3.0 0 o 0 3.0 ,0 o o o o 7 Results

Emergence trap collections A total of 17 species of Trichoptera were collected, 10 of these occurring in reasonable numbers. The collections of these ten species are summarised in Table 1 and the total collections for the years indicated grouped into 10-day catches, are shown in Fig. 1. Table 2 records the sex ratios for each species. Figure 2 illustrates the collections of T. waeneri during the whole period of collection and Table 3 shows the sex ratios of this species, arbitrarily divided into those that emerged in the early part of the emergence period (before 29th July) and

Table 2. The sex ratios and first and last dates of emergence of the commonest species at Llyn Hendref as recorded in the emergence trap.

Date Females Total trapped % First specimens Last specimens Tinodes 1963 315* 41 waeneri 1964 473 41 10/5 17/10 1965 279* 33 Ecnomus 1964 0 teneZlus 1965 17 59 28/6 29/7 Polycentropus 1964 50 44 1/5 18/9 jlavomaculatus 1965 11 45 1964 541 Cyrnus 32 14/5 14/9 trimaculatus 1965 186 31 Cyrnus 1964 25 8 30/4 29/7 jlci"vidus 1965 15 0 Limnephilus 1963 12 ? marmoratus 1964 51 47 26/5 8/8 1965 14 50 Anabolia 1963 25 32 nervosa 1964 78 45 19/8 24/10 1965 9* 11 Mystacides 1963 5 20 azurea 1964 17 47 19/5 25/9 1965 22 41 1964 5 60 Oecetis 9/6 7/9 lawstris 1965 38 24 Athripsodes 1964 6 50 26/5 29/7 aterrimus 1965 7 57 Trap in operation: 1963} 15/7-31/10 1964} 1/4-26/10 1965} 1/5-14/9 * Denotes incomplete collections. 119 ~ F('mol~ D Mole 8 A "~I va 100 '963& 1964 A Ot.f'lmV$ '964 & '96.5 6

WHiu l 10 15 20

120 120

100 100

80 BO

60

40 40 ...... " .

20 ...... ~ o [;;::: ::.. ... :h ~ .. Ol ~~~m~~.~.~.~.~.~.~.~~r :: : :.:: :::: ../: ~~ ..~.~.~~ Weo/I. I 5 10 15 20 1 5 10 15 10 I MA Y IJU Nfl JUI y I AUG ISEP T l OCI I IM .. yIJUNfIJUlyIAUC IHn lorr I Fig. 1. A summary of the emergence of ten species of Trichoptera from Llyn Hendref Anglesey for the years indicated. The data are grouped into the same lO-day intervals for eact year.

120 1964

qO 1965

10 o MALES #{,j FE M ALE S 1111-

;. ~t- u ~ ~Ol-

_ '0 0 iQ[ to. ;: ::f. 1" :J Z 10

11-19 19 -9 .-19 M •• J U l V Fig. 2. The emergence of Tinodes waeneri at Llyn Hendref as shown by emergence trap collections in 1963-1965.

121 Table 3. Tinodes waeneri. The number of adults that emerged into the emergence trap and in the laboratory arranged to show the sex ratio in each year. Emergence trap. Females Year Period Males Females Total % 1963 After 29th July 185 130 315 41.3 1964 Before 29th July 193 144 337 42.7 After 29th July 84 52 136 38.2 Whole 277 196 473 41.4 1965 Before 29th July 149 74 223 33.2 After 29th July 37 19 56 33.9 Whole 186 93 279 33.3 Laboratory emergences. Year Larval origin 1964 Llyn Hendref 39 33 72 45.8 Other Stations 6 7 13 53.8 Total 45 40 85 47.1 1965 Llyn Hendref 104 101 205 49.3 Other stations 17 23 40 57.5 Total 121 124 245 50.6 those that emerged later. Table 3 also includes the sex ratios of insects that emerged in the laboratory from larvae collected in the field. Figure 3 shows the fore-wing length of both male and female T. waeneri that emerged during the summer of 1964. Table 4 shows this data grouped into early and late emergences for each year and includes a comparison of males and females for each period as well as early and late insects of each sex for 1964 and 1965. Table 5 summarizes the total data for each year and compares the sizes (fore-wing length) of each sex in 1964 and 1965. Larval/pupal collections Each monthly collection was divided into the six developmental stages (5 larval instars and pupa) and the proportion in each stage is shown in Fig. 4. The lengths of the larvae were also recorded and provide a more detailed picture of the growth of the insects as is seen in Fig. 5. It is clear that considerable variation existed in the size of instar 5 larvae (Figs. 4 and 5) and the data for this instar alone are shown in Fig. 6. Records of water temperature at about 45 cm depth in Llyn Hendref are shown in Fig. 7.

Discussion

The ten common species of Trichoptera emerging from Llyn Hendref show three emergence patterns: (a) one discreet and relatively short period, the time of the

122 Table 4. Tinodes waeneri. Forewing lengths of adults caught in the emergence trap at Llyn Hendref. Statistical comparisons of males and females and early and late insects are included for each year.

Males Females No. in Range Mean±S.E. No. in Range Mean±S.E. Sample (mm) (mm) Sample (mm) (mm) 1963 After 72 5.25-7.87 6.743±0.058 67 5.77-8.40 6.818±0.061 d =0.89 29th July p>O.1 1964 Before 139 6.30-8.40 7.288 ± 0.033 177 6.82-8.40 7.636 ± 0.034 d =7.44 29th July P <0.001 After 72 5.77-7.35 6.741-0.044 46 5.77-8.19 6.750±0.072 d =0.12 29th July p>O.1 d =9.92 d = 11.26 p0.002 d =5.44 d = 3.54 p

~ N W Table 5. Tinodes waeneri. Summary of forewing lengths of emergence trap captures for all years and a statistical comparison of the sizes of the insects in the two complete seasons

Males Females

No. in Range Mean±S.E. No. in Range Mean±S.E. Sample mm mm Sample mm mm

1963 (part only) 72 5.25-7.87 6.743±0.058 67 5.77-8.40 6.818±0.061 1964 211 5.77-8.40 7.1D1±0.032 163 5.77-8.40 7.386 ± 0.044 1965 198 5.77-8.40 7.396 ± 0.028 110 6.30-8.40 7.574±0.046 1964/65 d =6.906 1964/65 d =2.948 p<0.001 p<0.01 >0.002

period varying with the species. Limnephilus marmoratus emerged between the end of May and early August. Ecnomus tenellus which occurred in 1965 only, emerged during the period late June to late July; Athripsodes aterrimus emerged between late May and late July; Anabolia nervosa emerged during September and October and Oecetis lacustris emerged mainly during June and July. (b) one prolonged period. Polycentropus flavomaculatus, Cymus trimaculatus and Tinodes waeneri emerged more or less continuously from early May to late September. (c) two distinct emergence periods. Mystacides azurea and Cymus flavidus both occurred in fair numbers during two well separated periods. The first group would seem to have only one generation per year whereas the third suggests two well separated generations. The second grouping, however, may also include species with two generations but these overlap each other. The two species of Limnephilidae show very different patterns of emergence and they agree with CRICHTON'S (1971) classification of flight periods. L. marmoratus was not recorded to emerge after August 8th in this study but the species occurred in CRICHTON'S light trap catches well into October (Wales and Northern England). This would be in line with the suggestion of an adult diapause in this species. Anabolia nervosa emerged between mid August and the end of October on Anglesey which corresponds almost exactly with CRICHTON'S data at the light traps and supports the classification of this species in the non-diapausing group. Of the other species captured, most emergence periods agree with HICKIN'S (1967) summary of flight periods. There is a discrepancy for E. tenellus which might be large enough to suggest the presence of a diapause and it is possible that C. flavidus may emerge later in southern England where the second generation may be longer. The captures of T. waeneri shown for each of the seasons studied (Fig. 2) showed some variation in emergence pattern with a strong suggestion of the existence of a second generation - particularly in 1963. Figure 5 and Tables 4 and 5 show that the insects decreased in size with the advance of the emergence period. An explanation of these phenomena is sought in the data on the occurrence of pupae and the size of the 5th instar larvae that gave rise to these. 124 80 lO ,. II f " ,:' , :( II 10 : I , I i' , '. I" 70 lO' ~I ' , , . .c II , , , ~ , , o!! tlI , , , " " :r " , . " , g- , , J 60 I ;i j ,. 1.1 ,. ,. I.II " , ,\1 20' , IS' JI

SO~----or----~~~r-__-'r---~-----r-- __,, __ --~ --__r- __-'rT __ -'-- __rT----' '0)0 )0 9 9' 19 19'9 '9 9 9 19 19'9 29- 1 1 - 18 II 11 21 7 J 11 17 17 June Jl,lly August Septtombll!!l Fig. 3. The range and mean fore-wing length of T. waeneri captured in the emergence trap during 1964. The size of each sample measured is indicated. Full lines = females. Broken lines = males. Figures 4 and 5 show the time of occurrence of pupae in each year and that these generally reflect the differences in emergence pattern referred to above. There could have been a clear separation of two generations in 1963 (when the emergence trap was not in operation during the early part of the period) but such a separation was not repeated in 1964 and 1965. It seems that the 5th instar larvae recorded in August are derived from eggs laid the same year and it is quite possible that some of these larvae pupated to give rise to a second (partial) generation of adults, CRICHTON (1960) and GOWER (1963) record such clear generations in southern England but it appears that the size of the second generation is variable in more northerly latitudes. Figure 6 shows that the 5th instar larvae occurring after the end of June-mid July were clearly smaller than those collected earlier in the season. This would undoub• tedly give rise to smaller adults than larvae pupating before the end of June. This may contribute to the decrease in size of adults referred to above.

l1li. Ii M J J . , J • J J • 1963 1964 Fig. 4. Tinodes waeneri. The relative distribution of the developmental stages present in each monthly sample.

125 ~ N II [J '",X{l I !ill!!!

~I! * , T ~-t- , l Tt t ; 1 1 1 , , ~' I 1'94, I "101 2Sl, 30 8 :U 9 2.4'tO J8 11 31" )91 IJl 1'S 0 1 90 J II I c:::J El rn

[t t Tt f" ... _ '" m $ , 1 76 , 113, '1, .1; '2oS, "2040 2)7 , . 8 28-9 ,6'0 2)'n "',, 190 .

I!: c.:::::J 8 01 IJ

II ~* f , ~; i I t I J"J'I n', n1j '10'. 1SI~ lOo 117 791 138 , . 0

1 '9 6 ~

Fig. 5. The length distribution of T. waeneri within each monthly sample of larvae. The number of larvae in each sample is included in the histogram and the proportion of pupae present is shown stippled. 12·0

11·0

10·0

f" ~ 56 22 1U 70 1 '~.11""" 89 l1'70 51 ~ 99 1 46 12 28 28 64 34 36 60 36 6" 78 32 42 29

50 1 M SON OIJ F M A M J A S 1965

Fig. 6. T. waeneri. The mean length, range and Standard Deviation of fifth instar larvae collected in each monthly sample. The size of each sample is indicated.

~ N -..J I-" tv 00

26

24 1

: . I T l"1 "T T

ll1 i:· 1j 11) r ·rrl llip. 1111

o 1A r : r J r J r A r S*;O r N r I ~r F r M r A*'r M r J r J r A r : r 0 r N r 0 I J I F riM r A "{l-M r J r J ' A r S r 0 r 1963 1964 1965 Fig. 7. Llyn Hendref. Water temperatures recorded in about 45 cm water. The maximum, minimum and the temperature at the time of reading are shown. * Indicate the time of appearance and disappearance of T. waeneri pupae. The explanation for the difference between the years is obscure but may well be linked directly or indirectly to temperature and/or light intensity. These factors affect larval growth (directly and through food supply) and pupation but I have no detailed records which will explain the observed -differences. Records of hours of sunshine at a nearby meteorological station showed little difference between the years (June• August totals = 537, 531, 593 hours respectively) - being particularly similar in 1963 and 1964. Water temperature records did show some differences between years (Fig. 7) but again it is difficult to see how these could explain the differences in emergence patterns that were recorded. In conclusion, it would appear that T. waeneri has a variable emergence pattern which may be one of the reasons for its ubiquity. It is also suggested that significant differences can exist in the size of the adults depending upon their time of emergence, within any year, as well as between years. There are, therefore, likely to be considerable differences in size between specimens from different localities.

References

CRICHTON, M. I. 1960. A study of captures of Trichoptera in a light trap near Reading, Berkshire. Trans. R. ent. Soc. Lond. 112(12): 319-344. ,--. 1971. A study of caddis flies (Trichoptera) of the family Limnephili(tae, based on the Rothamsted Insect Survey, 1964-68. J. Zoo!., Lond. 163: 533-563. GOWER, A. M. 1963. A study of the biology of Limnephilus lunatus (CURTIS) and Drusus annulatus (STEPH.) in watercress beds. Unpublished M.Sc. Thesis: University of Reading. HICKIN, N. E. 1967. Caddis larvae. Larvae of the British Trichoptera. London: Hutchinson xii+476. REYNOLDSON, T. B. 1958. The quantitative ecology of lake-dwelling triclads in northern Britain. Oikos 9: 94-138.

Discussion

HILEY: It is a general rule not only in insects but in other organisms such as fish that animals which have a long growth period under cool conditions reach a larger size than animals which have a short growth period under warm conditions. In our studies on chalk streams we have used this very fact to determine the life histories of some Ephemeroptera as there was no other way to determine them. We found very clt:ar differences in size even though there was sometimes a long spread of emergence which included overlapping generations. The second point is with regard to your results from max-min thermometers. Although you have not shown any change from year to year this need not exclude the possibility of temperature being operative. If you tried using cumulative temperature records you may yet find a correlation. JONES: Ideally, continuous temperature recordings are required so that the cumulative effect can be quantified. Temperature may also act through the food supply. HILEY: What do they feed upon? JONES: On algal scrapings from rocks etc. HILEY: Have you tried feeding the larvae on wheat? This is very good for some species, e.g. Leptoceridae larvae. JONES: I fed early instar larvae on yeast.

129 HILEY: Try wheat. You can grow a great many species very fast on it, and if you break the wheat grains it is possible to feed very small larvae this way. Even some leptocerids, which are supposed to be algal scrapers, grow very well on wheat in the later stages (e.g. Mystacides, Athripsodes cinereus, A. aterrimus). RESH: In Hickin's text, Caddis Larvae, he reports that the larva of Ecnomus tenellus is associated with freshwater sponge. Have you ever observed this association? Has anyone else ever observed this species in sponge? HILEY: I haven't seen it in sponge, but I have seen some Tinodes and Lype larvae in twigs and branches, and it is possible that their tunnels would pass accidentally through the sponge which grows in the same situation. NIELSEN: I think that the association of Ecnomus tenellus with sponge is quite accidental. It is a net-spinning species, and I shall take the opportunity to say some words about the systematic position of the subfamily Ecnominae. In many textbooks it is now included in the family Psychomyidae. However, both as to the morphology and the ecology of the larvae and the morphology of the female genitalia they absolutely are polycentropodids. CRICHTON: On Ecnomus tenellus I could add a detail which I did not mention in my talk. The light trap records for southern England show a second smaller peak in late summer. HIGLER: I caught it very irregularly in several places as you found it only once in three years. I found it now here now then, very irregular and throughout the summer, quite a longer period, than 3 months.

130 Proc. of the First Int. Symp. on Trichoptera. 1974. Junk, The Hague

Studies on the eggs, larvae and pupae of Tinodes waeneri (L.)

N. V. JONES

The larvae of Trichoptera have attracted the attention of many workers but the eggs and pupae have been largely ignored. This paper describes observations and experi• mental studies made on the eggs and larvae and, to a lesser extent, the pupal stage, of Tinodes waeneri. This work formed part of a broader study of the ecology of some psychomyiid caddis-flies (JONES, 1967).

Methods

Larvae and adults of T. waeneri were collected from Llyn Hendref, Anglesey (SH 398765) and maintained in the laboratory. Larvae were kept in covered dishes and provided with sand for building their galleries and natural organic sediment for food. If they were kept in a lighted situation it was not necessary to add food very often as some algal growth occurred naturally. Larvae kept in this way survived for long periods and many pupated. If this occurred on the sides of a glass vessel, it was possible to observe the changes taking place during the process of metamorphosis. The adults that emerged from such bowls could be collected and their age known to within 12 hours or less. The imagos normally lived for about 7 days if provided with water and they would readily lay eggs on damp filter paper. Their life could be prolonged up to about 14 days by the provision of a weak sugar solution. Egg masses could easily be divided into egg groups for experimental purposes. The eggs were maintained on wet filter paper or on glass slides kept in small chambers above solutions of various salts to provide a range of Relative Humidities as outlined by SOLOMON (1951).

Results

The eggs The females of T. waeneri usually deposited their eggs within six days of emerging (Table 1). Under natural conditions the eggs were laid at dusk on solid substrata at the water surface where they formed encrusting masses. Sometimes consecutive

131 Table 1. T. waeneri. Age records of females laying eggs in the laboratory Age at laying (Days) Number recorded

1 2 2 3 3 4 4 1 5 6 6 2 7 1 8 1

Total 20

layings could be seen as egg lines above the water surface if laying took place during a period of falling water level (Fig. 5). Eggs laid in such situations are clearly vulnerable to changes in air temperature and humidity and so some experiments were carried out to investigate this aspect of their ecology. The eggs are yellow ovoid bodies enclosed in a transparent jelly. Their size varied between egg masses but lay within the range O.21--O.26xO.16--O.19 mm.

Effect of temperature on development Eggs laid in the laboratory and, therefore, of known age were kept in water at a range of constant temperatures and their development followed. Figure 1 shows the results of the experiment. Development was arrested at 2 °C and proceeded only

25

20

~ 15 .= .0- i :j I

0~1~,--~,10 15 ~'~'C-T'~'--~'~'~~' 20 25 30 35 40 45 50 ~~,---~~s's 60 6'5 7~

Developmental period (days) Fig. 1. T. waeneri eggs. Developmental period under a range of constant temperatures.

132 slowly at 6°C when only eggs laid two days before the start of the experiment actually hatched. New laid eggs showed development but for some reason, failed to hatch at this temperature. As constant temperatures are unlikely to be encountered in the field for any period of time a second experiment was carried out when eggs were exposed for various periods of time to temperatures likely to be experienced in nature. Figure 2 shows

9

8

7

6

'p... 5

~ 4 ::s III o Q. )( w 3

2

O+------.-----,r-----~----_r----_,------~----~----~--- 7 8 9 10 11 12 13 14 15 Developmental period (days). Fig. 2. Graph showing the effect of different exposures to 9°C on the developmental period of eggs. Line drawn by eye.

133 Table 2. T. waeneri. The results of exposing eggs to several periods at different temperatures (a) New laid eggs (less than 1 day old)

Temperature Control IS-23"C 24°C 15 °C 9°C 6°C

Exposure period days Whole period 2 Whole 2 Whole 2 Whole

% Hatched 100% 100% 100% 80% 100% 100% 80% 100% 80% 100% 0

Development time in (50% hatch) days 8 6 8 9 14 8 10 32 9 II

(b) Partly developed eggs (2 days old)

Control 15-23°C 24°C 15°C 9°C 6°C

Exposure period days Whole period 2 Whole 2 Whole 2 Whole

% hatched 100% 100% 90% 100% 90% 100% 100% 100% 80% 100% 100%

Development time 8 7 8 9 12 9 II 32 9 11 70

the effect of exposures to a temperature of 9°C on the development time of eggs from the same egg mass. It is clear that the eggs could withstand developmental arrest by low temperature and could continue normal development subsequently. This also applied at 6°C as can be seen from Table 2 which shows the effect of exposure for one and two days to various temperatures on the development of both new laid and partly developed eggs.

Effect of humidity on development The results of exposing eggs from the same egg mass to various humidities is shown in Figure 3. All eggs kept at 58% RH. were dead after one day and only a few survived two days at 84% RH. The eggs in the 100% R.H. chamber hatched on the glass slide but this did not happen at 98% R.H. This was attributed to the water that had condensed on the slide in the former case and which in some way facilitated the hatching process. The eggs were clearly vulnerable to desiccation as three days at 92% RH. killed 75% of them and they could hardly survive one day at 84% RH. The larvae T. waeneri larvae have been described by HICKIN (1950) and LEPNEVA (1970) and the species features in EDINGTON and ALDERSON'S (1973) key to the British psychomyiid larvae. I add Table 3 and Figure 4 which record the sizes of the heads of

134 Table 3. T. waeneri. Distinguishing features of the five larval instars.

Head width Number of large Range Mean±S.E. Number Instar bristles on prolegs (mm) (mm) sampled

I 2 0.147-0.196 0.161±0.001 75 II 3 0.233-0.269 0.252 ± 0.002 18 III 4 0.331-0.429 0.376 ± 0.002 73 IV 5 0.453-0.588 0.532 ± 0.004 56 V 6 0.711-0.845 0.764±0.004 58 the five larval ins tars as well as the relationship of larval length with ins tar. The number of large bristles on the anal prolegs was found to be characteristic of ins tars as recorded by DANECKER (1961). Substratum selection The selection of substrata shown below was offered to 20 fifth instar larvae and silt and sand added. The distribution of galleries built within 24 hours was as follows: Distribution of larval galleries in three replicates Coarse Rough gravel stones 6 I 5 10 I 4 2ll Smooth Silt and 7TT 5TO 710 stones sand only (a) (b) (c) (The departure from a total of 20 is due to some larvae not building.)

The pooled results were subjected to the Chi2 test which showed that a significant selection of substratum had taken place (X 2 = 23.6 P < 0.001). The smooth bottom with fine sediment had less attraction than the other substrata and more larvae built in the gravel than on the stones. The larvae using stones built between these and the bowl. Larvae were capable of building on smooth polythene as it was usual to keep them in such bowls. Building material selection The selection of material for building the larval galleries was also investigated by a simple choice experiment. The choices offered and the results are shown below for both fourth and fifth instar larvae (20 larvae were used in each case). Distribution of larval galleries in duplicates 1.0 mm Clean 5th instar 00 0 $2 2 $2 3 0.5 mm <0.1 mm 5 I 8 0 4th instar 0 ~O $4 9 I 14 3 I 2 0

135 A

100 " 25

~

75 20

SO 15

15 __ ~ _ _ •__ .A _ _ _ _ • - _ . - - - -; 10

0

B C 100 25 "

~ 75 20

.: 50 15 _0- - -; -- \ 25 ~ ~ - 10 ~r \ \ i ~ \ ~ \ ~ '" \ 0 .., . ~ c ~ II. ~ v lO O D E 25 ;;

~ "E (I. Co Q 7S 20 0; > ~ 0

SO 15 .- 25 10

0 0 6 8 9 0 6 8 EJlpo$u,e In days

136 26

24

22 ~ 3 20 ill - ,

'8

16 , 6 7 9 10 11 Larval length (mm)

:: 12

'0 10

; 8 II

= 6 Z

20 3 a 40

Head width In gratlcule diVISions Fig. 4. The head widths and larval lengths of T. waeneri related to the five instars.

When the figures for the larvae selecting the finest sediments (0.1 mm and <0.1 mm) are pooled and the X2 test applied to the four selections thus made, it is found that the two instars distributed themselves quite differently (X2 20.46 P < 0.001). The fifth instar larvae selected for the 1.0 mm material and against the 0.1 and 0.2 mm particles (X2 = 10.8 P < 0.05 > 0.001). The fourth instar larvae showed a more significant selection (X2 = 22.6 P < 0.001), being particularly attracted to the 0.2 mm particles and to a lesser extent to the 0.5 mm material. The larvae used their mandibles to carry material and it is suggested that the majority of particles used by a larva will be related to its mandibular gape. Both larger and smaller particles can be used by the use of the head to push, or the legs to manipulate them. Reaction to light The offer of dark or light areas for building showed that late instar larvae did not select with respect to this parameter. This is in agreement with DANECKER'S (1961) observation on T. zelleri. First instar larvae showed a positive light reaction when released in a dish but it was not clear whether this affected their choice of settling sites.

Fig. 3. Graphs showing the effect of a range of different Relative Humidities on the develop• mental period and percentage survival of T. waeneri eggs. (A) 100% R.H.; (B) 98% R.H.; (C) 92% R.H.; (D) 90% R.H.; (E) 84% R.H.; Full line = Percentage of eggs hatching; Broken line = Developmental period.

137 Fig. 5. Lines of Tinodes waeneri eggs laid during a period of falling water level. Reaction to current Larvae were introduced into a sloping channel which had a sand-impregnated bottom and down which water flows of up to 40 cm sec-1 were achieved. Last instar larvae showed the following reactions:

Current speed Reaction

35 em sec-I Larvae usually washed out in a short time. 27-30 em sec-I Larvae wandered about but did not build. 15-20 em sec-I Larvae built galleries. If current subsequently increased to 30-35 cm sec-I, the larvae evacuated the galleries. 15-25 cm sec-I Some larvae pupated and adults successfully emerged.

Fourth ins tar larvae were less able to cope with currents above 20 cm sec- 1 • Larval diet The food of the larvae was found to be the organic and inorganic deposit present on the stones on which the larvae lived. The only selection noted was of the size of particles ingested. Feeding took place at the end of the galleries where the larvae scooped deposit with their mandibles. This often resulted in areas at one or both ends of the gallery being devoid of deposit. The extension of the gallery was often dictated by the availability of food. These observations confirm several published records, e.g. AHMAD (1939), SATIJA (1957), DANECKER (1961).

138 The pupa Metamorphosis was not examined in detail but during studies on the duration of the process it became clear that three phases were distinguishable:

1. The preparatory stage (Fig. 6). The larva constructed its pupal case which was shorter and more robust than the larval gallery. The larva became shortened and fat and became less active with the head assuming the hypognathous position (as opposed to the larval prognathous position). The legs became progressively inactive in the sequence meta-, meso-, and prothoracic. The abdomen was apple green and performed respiratory undulations like the larva. 2. The green pupa (Fig. 7). The larval exuvia was shed and pushed to the posterior end of the case. The pupa thus formed had a green colour which was gradually lost and the eyes changed colour from brown to black. The change from larva to green pupa was quite abrupt and was accompanied by a change in respiratory movements. The abdomen was now still and water drawn through the pupal case by the lunging movements of the head and thorax. 3. The brown pupa. This stage developed gradually from the former but was identifiable by the assumption of the adult-like colour. This stage did not change much in appearance until the adult emerged from it (MORTON 1890; HICKIN 1950).

Fig. 6. Larva building pupal case.

139 Fig. 7. The green pupal stage.

The duration of these stages was recorded as follows under laboratory conditions (Temperature 17-22°C);

Number Whole Sex observed Preparatory stage Green pupa Brown pupa pupation period

Male 9 5.0 7.2 3.6 15 .99 days Female 16 4.8 6.8 3.5 15 .25 days

The complete pupation period was recorded for 189 insects under laboratory conditions and the data are summarized as follows.

Number Mean pupation period Sex observed days ± S.D.

Male 91 15 .7±1.6 [ d = 2.954 ] Female 98 15.1 ± 1.4 P<0.01>0.002

The effect of temperature on the length of the pupation process was only examined at the two extremes likely to be encountered in summer in British waters. At 24°C metamorphosis was completed in 12 days, whereas at 9°C it took from 40-48 days.

140 It was estim~ted that under field conditions in North Wales metamorphosis would require about 15-18 days.

Discussion

The females of T. waeneri were not mature on emergence from the pupal stage but they became capable of laying eggs within a week. The eggs were laid in situations where they were particularly vulnerable to desiccation. Under normal conditions they would be kept moist by wave action or high humidity close to the water surface but they could be left dry by the falling water levels during drought situations or on the shores of regulated lakes. The eggs were capable of withstanding developmental arrest brought about by drops in temperature of either short or relatively long duration. Observations of larval behaviour largely confirm earlier work, particularly that of DANEcKER (1961) who carried out very interesting experiments in the natural situation on T. zelleri larvae in hygropetric situations. The larvae seem able to build 'on any substratum as long as building material is available. They can utilize a range of materials but the size predominantly used is suggested to be related to their mandibular gape, as they carry particles with their mandibles. T. waeneri larvae were capable of completing development in a water flow of up to 25 cm sec-1 but would seek another site when the current became faster. This might be expected of this species which is characteristic of lake shores and slower reaches of streams. The stages of metamorphosis seem particularly interesting and the use of these larvae that pupate on glass offer a way of examining the process in more detail. The obvious change in respiratory behaviour accompanying the transition from larva to pupa was striking and must be due to the radical changes of musculature taking place at that time.

References AHMAD, N. 1939. The mouth parts and alimentary canals of trichopterous larvae in relation to their feeding habits. Ph.D. Thesis: University of London DANECKER, E. 1961. Studien zur hygropetrischen Fauna. Biologie und Okologie von Stactobia und Tinodes (Trichoptera). Int. Revue ges. Hydrobiol. Hydrogr. 46(2): 214-254. EDINGTON, J. M. & ALDERSON, R. 1973. The taxonomy of British psychomyiid larvae (Trichop• tera). Freshwat. BioI. 3: 463-478. HICKIN, N. E. 1950. Larv~e of the British Trichoptera. 28. Tinodes waeneri L. Proc. R. ent. Soc. Lond. (A) 25(7-9): 67-70. JONES, N. V. 1967. The ecology of some psychomyiid caddis-flies. Ph.D. Thesis: University of Wales. LEPNEVA, S. G. 1970. Fauna of the U.S.S.R. Trichoptera I. Larvae and Pupae of Annulipalpia. Translation from 1964 Russian edition by Israel Program for Scientific Translations. Jerusalem. MORTON, K. J. 1890. Notes on the metamorphosis of two species of the genus Tinodes. Entomologist's mono Mag. Ser. 2, vol. 1(26): 38-42.

141 SATIJA, G. R. 1957. Studies on the structure and biology of certain Trichoptera larvae with special reference to food and feeding habits. Ph.D. Thesis: University of Durham. SOLOMON, M. E. 1951. Control of humidity with potassium hydroxide, sulphuric acid, or other solutions. Bull. ent. Res. 42(3): 543-554.

Discussion

NIELSEN: I have made some unpublished observations on the building actIvity of Tinodes waeneri. The larva builds a several centimeters long vault of fine sand grains. It feeds on the covering of blue green algae on the stones. When it has 'grazed the stone off' as far as it can reach without entirely leaving the building with its anal prolegs, it moves the building, breaking it down at the posterior end and carrying the material forward within the building. Though the two ends of the building look alike, one is functionally the anterior end. The building is initiated in some angle between a sandy bottom and a stone. FLINT: You say you kept the eggs under varying humidities. Were these eggs still in their gelatinous covering, or were they separated from their gelatine? JONES: The gelatinous covering was still on the eggs. CIANFICCONI: Have you seen any parasites on Tinodes waeneri? JONES: I was not looking for parasites but some adults were seen with mites under their thorax. CIANFICCONI: Are there lakes where T. waeneri and T. pusillus live together? JONES: Only waeneri. as far as I can make out. CIANFICCONI: Have you noted whether the specimens which emerge in September-October have setal differences in the pubescence compared with those who emerge in summer? JONES: No, I did not notice any such difference. VAILLANT: Many years ago in the hydro biological institute of Grenoble there were several large tanks 4 to 5 feet deep, with flowing water; hundreds of big rainbow trout were kept in them all the year round. All summer the walls of the tanks were covered with Tinodes waeneri larvae. The trout did not try to eat them. Some trout were opened and very few larvae of Tinodes were found in their stomachs. I wonder why? JONES: I found only adult T. waeneri in fish stomachs. The imagos are also eaten by spiders. The larvae are taken by Erpobdella octocu/ata. V AILLANT: On one occasion in early spring, all the tanks were cleaned and left to dry a few days. Nevertheless, the following autumn, some imagos of Tinodes waeneri emerged. Wouldn't that mean that there are two generations a year? Ross: In some Trichoptera, you get continuous generations during the summer with different sizes of larvae present in early winter. The larger ones pupate but the smaller ones seem to continue growth and they catch up. The species then has one mass emergence in spring. JONES: It is a possibility but I have no evidence of it. HANSELL: It has been recorded for a number of species of larvae with free moving cases that selection of building materials is usually determined by manipulation of the particles, sometimes very prolonged, with all the legs. The function of this appears to be to separate suitable from

142 unsuitable building material on the basis of shape and possibly size. In the method you describe the particle size has an upper limit clearly set by the size of the gape of the mandible. I wonder if you noticed any mechanism which might set a lower limit to particle size acceptance. Do larvae take up some particles in their mandibles, test them and reject them? JONES: There must be a limit below which manipulation by a pincer movement of legs or mandibles becomes difficult. My experiments were restricting in that the larvae were building in material of one size group.

HANSELL: The main evidence for the selection of particle size must be that firstly the house is composed of a size distribution of particles which is different from that found in the suhstrate, and secondly whether you observe that larvae take up some kinds of particles and, having manipulated them, drop them. Ross: What is the position of the pupal case in relation to the tube?

JONES: It is normally built within the tunnel, the latter usually disintegrates. I do not think that it normally bears a fixed relation to any particular part of the tunnel. FLINT: Do you find that the pupae are orientated in any way in relation to the water surface?

JONES: They are usually horizontal.

143 Proc. of the First Int. Symp. on Trichoptera, 1974, Junk, The Hague

A progress report on the North American Macronema larvae: their retreats, food and feeding nets (Trichoptera: Hydropsycbidae).

JAMES BRUCE WALLACE

The structure of the larval retreats of Macronema transversum WALKER, M. carolina BANKS, and M. zebratum HAGEN are described. Larvae of all three species each construct characteristic dwellings. Those of M. transversum are found on rocks in large streams and probably represent one of the more complex feeding and dwelling structures found in the non-social insects. The retreats of M. carolina are constructed in bark and outermost wood of submerged tree limbs and those of its sister species, M. zebratum, are constructed of sand grains on the sides of rocks. The feeding nets of mature larvae have very fine meshes with rectangular openings ranging from 3.3 x 28.6 to 5 x 40 fLm. The minute meshes of the net are used to filter microseston (fine particulate organic matter, phytoplancton and bacteria) upon which the larvae graze. Larvae of all three species possess morphological structures such as densely pilose brushes on the fore tibiae, labrum, labium and oral opening which are used for grazing on minute food particles trapped by the net. Evidence is presented that Holarctic choice of habitat of Macronema spp., i.e. large streams, may be influenced by microsestonic feeding habits of their larvae, rather than strict water quality parameters often used by aquatic biologists. The comparative retreat building behaviour of known Macronema species is discussed. (Read by JOHN C. MORSE.)

Discussion

RESH: I understand that in the past few months a group of civil engineers have proposed that certain heavily polluted streams be 'seeded' with net-spinning hydropsychid larvae, in the hope that these caddis-flies would filter and remove the suspended particulate matter from the water. The problem with a program such as this is that there is usually a reason why a species is not occupying a particular habitat, and I doubt that 'seeding' an area with these species would ever establish a permanent population.

BADCOCK: Are the nets and retreats always on the upper surface of the stones where they are exposed to the flow of water and getting the most benefit from it, or do you find them underneath?

MORSE: The slides illustrate that the nets do occur at various places on the stones. Of course the opening of anyone must be in the flow, even if at different sites on the stone.

145 HILEY: From the results of recent studies on chalk streams it has been found that the detrital material which is in suspension may be of a very different quality, in terms of food value, to the material which is rolling along close to the bottom of the stream. Much of the material on the bottom of the stream has recently been eaten, and is often in the form of fecal pellets. The material which is in suspension may have come from broken fecal pellets, and probably has a new bacterial and fungal flora which would make it a better food resource. The position of the opening of the net may be determined by the quality of the food. BOURNAUD: Are the larvae affected by pollution, they are as filtering perhaps good indicators for pollution? MORSE: I suspect that this is very true. Hydropsychids are sometimes considered as indicators of various sorts of pollution. The idea that Dr. WALLACE presented here is simply that Macronema species may be studied as indicators of the presence or absence of suspended detritus particles of a certain small size. FLINT: The Potomac River was just such an area. At one time, it supported transversa, we have old material in our collection, but it has not been seen in years. I think silt, which would clog the very fine nets of this species could be the reason that transversa and perhaps carolina are no longer found in our area. BADCOCK: Is there a seasonal factor in the spinning of the net? Is there a temperature factor involved, for instance? Also, does Macronema always use its nets for feeding, or may it browse on the bottom or seize small organisms if it is not spinning a net? MORSE: I cannot answer this. I have not worked with the ecology of these species myself.

146 Proc. of the First Int. Symp. on Trichoptera, 1974. Junk, The Hague

The interpretation of light trap catches of Trichoptera from the Rothamsted Insect Survey

M. I. CRICHTON

Introduction

The Rothamsted Insect Survey is based on a network of light and suction traps in Britain. Its purpose is to study changes in the insect population, with particular reference to pest species. TAYLOR (1974) describes the techniques used and reviews the present state of the survey and some of the results which are emerging. The light trap part of the survey was started in 1964 as a record of Macrolepidopt• era from the standard Rothamsted Light Trap (WILLIAMS, 1948) with a 200 w tungsten filament lamp, operated throughout the year. In 1965 seven traps com• pleted a whole year's sampling. The number of sites has grown each year and by the end of 1973 there were 160 traps in operation. Caddis-flies are the only insects, other than Lepidoptera, which have been re• corded systematically from the light traps over a number of years. They have been sent to Reading for identification and recording. An analysis of the catches of Limnephilidae for the years 1964-1968 has already been published (CRICHTON, 1971). This communication deals briefly with Limnephilidae and other families over the years 1965-1971. Detailed information and analysis of the records will be published elsewhere.

Light trap sites

Caddis-flies have been recorded from seventy-five sites in many different habitats. Few of the traps were in the immediate vicinity of water, but there were often streams in the neighbourhood. In this study use is made of sixty-seven sites from which a total of 148 complete year records have been obtained. The location of these sites is shown in the map of the British Isles (Fig. 1). For the purpose of analysis they are grouped in the following three geographical areas: A. Scotland, 15 sites. B. Wales and Northern England, 18 sites. C. Southern England, 34 sites.

147 o Miles 100 I ,

I I i o Km 100 o

Fig. 1. Map of the British Isles showing the location of light traps in the Rothamsted Insect Survey from which complete year records of caddis-flies were obtained. The three geographical areas are: (A) Scotland; (B) Wales and Northern England; (C) Southern England. (Because of the small scale of the map not all the sites in area (C) could be shown as separate dots.)

148 Composition of catches

The Hydroptilidae were excluded from the study because of the difficulty of separating them from other small insects. This leaves 165 species of caddis-flies. A comparison of the number of species of Limnephilidae with those belonging to other families is given in Fig. 2. The higher proportion of Limnephilidae can be explained by their wide dispersal by flight and frequently long adult life, which may include a diapause. They were thus often caught far from water. Because of their weaker flight and shorter life span, species from other families were poorly rep• resented, especially in traps away from water. These families included more rare and local species which were not caught, while many of those recorded were present in small numbers perhaps because they flew more by day than by night. There were also six common and widely distributed species which were not caught in any light trap, because they appear to be predominantly day-fliers. These six species were:

Agapetus fuscipes CURT. Athripsodes nigronervosus RETZIUS A. albifrons L. Triaenodes bieolor CURT. Silo nigricornis PICf. Brachycentrus subnubilus CURT.

It should be noted that there were no limnephilids in this category. At most sites the composition of the catch in number of species and individuals was closely similar in successive years, as is illustrated for six sites in Fig. 3. This feature is especially marked in the Limnephilidae. The light trap at Dale Fort, in south-west Wales, was in an exposed position close to the sea without any freshwater habitats in the vicinity: the catch was small and consisted entirely of limnephilids. In contrast, the other sites in Fig. 3 had streams in their vicinity and so families other than Limnephilidae were represented.

Flight periods

Some species of Limnephilidae have a flight period extending from early summer until late autumn. The significance of this feature was not understood until NovAK & SEHNAL (1963) showed that the early-emerging females had immature ovaries which developed in the shortening days of autumn after passing through a. summer diapause. The maturation period could be extended by keeping the flies under long-day conditions. This pattern of flight period was confirmed by CRICHTON (1971) for several species collected in the Rothamsted Insect Survey. The catches of Limnephilus affmis CURT. are given here as an example (Fig. 4), which also shows the characteristic found in other limnephilids of a median week which is progres• sively later in the south. In Scotland the record of this species shows no early

149 TRICHOPTERA IN ROTHAMSTED INSECT SURVEY, 1965-71

British species 194 Hydroptilldae 29 (not included) 165 ~l Limnephilidae Other families % 56 109 100 ...... 6 common. 4 ...... day-fliers

15 rare

80 not recorded

26 local

60

40

62 recorded

20

OL-----~~~--____~LL~ ______

Fig. 2. A comparison of the representation of species of Limnephilidae and other families in the light traps. 150 Ind~ ,------, 1600 Fort Augustus Ardross r- M~ntwrog - - - 1200 spp - 40 800

20

'67 '68 '69 '66 '67 'S8 67 58 69

2000 Wisley Ind~ ,..-

Dal~ Fort 80 1600

N~tllecombe

60 1200

spp - 800 spp. I 10 20 20 400

65 'S6 '67 'S7 '68 '69 '68 69 70

01 her famlllH ] Other om Ill eS] o spec i es o IlldlVlduols I l,mn.ph,l,da. ~ Limnephilida.

Fig. 3. Total catches for three years at Ardross and Fort Augustus in Scotland, at Maentwrog and Dale Fort in Wales, and at Wisley and Nettlecombe in Southern England. emergence peak so it may not pass through a summer diapause as in England and Wales. Another well-defined pattern of Iimnephilid flight period (type 3 in CRICHTON, [971) is illustrated by Halesus radiatus CURT. in Fig. 5. Here the adult flight period s restricted to the autumn and there is therefore no summer diapause. As with other imnephilids, there is a shift to a later median week in the south as compared with .he north of Britain. In families other than the Limnephilidae a common pattern is a short summer 151 20 30 40 weeks

0/0 Limnephilus 30 affinis

20 A 182

10

% 20 B 353

10 ,J,

%

20 C 452

10

Apr.

Fig. 4. Weekly catches of Limnephilus affinis expressed as percentages of the recorded total catch for each area. The arrows indicate median weeks. flight period with a well-defined single peak. Examples of this kind from the light trap catches are: Rhyacophilidae, Agapetus delicatulus MeL. Leptoceridae, Athripsodes dissimilis STEPHENS Lepidostomatidae, Lepidostoma hirtum F. In contrast to the above there are species with two peaks of activity in some or all 152 30 40 weeks

0/0 Halesus radiatus 20. A 854

10

0/0 20 B 785 - 10 -

0/0

20 c 290

10

Jun·1 Jul. I Aug. I Sep. Oct. I Nov.

Fig. 5. Weekly catches of Halesus radiatus expressed as percentages of the recorded total catch for each area. The arrows indicate median weeks.

of the geographical areas, suggesting more than one generation in the year. Thus Tinodes waeneri L. in Fig. 6 has two peaks in the south of England. In northern areas there are single peaks and shorter flight periods. On the other hand JONES (1975) has found evidence of a partial second generation in Wales from collections of larvae and pupae. It must be emphasized that light trap catches give only an indication of life cycle patterns; confirmation must come from regular collections of larvae and pupae throughout- the year.

153 Distribution maps Density distribution maps were produced at Rothamsted for a number of species, using the Symap V Program of the Laboratory for Computer Graphics, Harvard. They are based on the average annual catch over the period 1966-1971 for 64 sites in Britain. The layering intervals are approximately geometric and the levels are represented on the maps by symbols which give a progressively denser effect for higher numbers. Examples of density distribution maps for Hydropsyche angustipennis CURT. (Fig. 7) and Lepidostoma hirtum F. (Fig. 8) are reproduced here. BADCOCK (1975) has 20 30 40 weeks % Tinodes waeneri 20 A 495

10 -

% 30 - B 328 20

10

% 20 r- C 2077 .J,

10 r- r May , Jun. , Jul. , AUg., Sep. I Oct.

Fig. 6. Weekly catches of Tinodes waeneri expressed as percentages of the recorded total catch for each area. The arrows indicate median weeks. 154 .. "":."" .. .. "" ... :

: .. ... ::::.

"";,';';;!;,, 'i,i :1!ijllllllf;' iil;!li!~!!I!I;:ij, !;ilili '::,"' ;~';;'.:,:'" j:i!li~~li;! j; '::" i

Fig. 7. Density distribution map for Hydropsyche angustipennis, based on the average annual catch in the period 1966-1971. The approximately geometric layering intervals are 0,1-2,3-9, 10-31.

155 · ...... ~

Fig. 8. Density distribution map for Lepidostoma hirtum, based on the average the period annual catch in 1966-1971. The approximately geometric layering intervals are 0,1-2,3-9,10-31, 32-99, 100-315, 316-999, 1000-3161. 156 shown from collecting and a study of published and museum records that H. angus• tipennis is associated with the lower reaches of rivers and is more common in the Midlands and the south of England. In Northern Britain it is known from a few streams which are warmer than others in summer; she records it from only four sites in Scotland. There is good correspondence between her map produced by the Biological Records Centre and the density distribution map from Rothamsted. The association of L. hirtum with upland areas is clearly brought out in Fig. 8, and is in contrast to the distribution of H. angustipennis.

Acknowledgements

I am indebted to Dr L. R. TAYLOR for allcwing me to use the Rothamsted Insect Survey organization, and to many helpers, particularly Mrs JOAN NICKLEN, for separating out caddis-flies. I must also thank Mr I. P. WOIWOD for handling the computer side of the work at Rothamsted. I am especially grateful to Mrs DOROTHY FISHER who carried out the major part of the tedious work of identifying thousands of caddis-flies and also helped in many other aspects of the study. She was supported at Reading by a grant from the Natural Environment Research Council. References BADCOCK, RUTH M. 1976. The distribution of the Hydropsychidae in Great Britain. Proc. I. Symp. Trich.: 49-57 CRICHTON, M. I. 1971. A study of caddis flies (Trichoptera) of the family Limnephilidae, based on the Rothamsted Insect Survey, 1964-1968. J. Zool., Land. 163: 533-563. JONES, N. V. 1976. The Trichoptera of the stony shore of a lake, with particular reference to Tinodes waeneri L. Proc. I. Symp. Trich.: 117-129 NovAK, K. & SEHNAL, F. 1963. The development cycle of some species of the genus Limnephilus (Trichoptera). Cas. cesk6 Spa!. ent. 60: 68-80. TAYLOR, L. R. 1974. Monitoring changes in the distribution and abundance of insects. Rothamsted Experimental Station. Report for 1973: 202-239. WILLIAMS, C. B. 1948. The Rothamsted light trap. Proc. R. ent. Soc. Land. (A) 23: 80-85.

Discussion

NIELSEN: I am surprised by the last map because Lepidostoma hirtum was - prior to the extensive pollution - very common in Denmark, which, as you know, is a lowland. Is there any absence of lakes in this part of Britain? CRICHTON: We must not deduce too much from these maps because the light traps of the Rothamsted Insect Survey are set up in many different habitats with only a few in the immediate vicinity of water. JONES: I am a little bit surprised by your statement that you find roughly the same numbers from year to year. I expected them to be a little bit more variable. Does this in fact apply to other groups of insects caught in this way? CRICHTON: The relative abundance of a species at anyone site is remarkably constant from year to year, although the actual numbers do of course vary. In general this applies to the Macrolepidoptera, the study of which is the main object of the survey. 157 HILEY: In some groups of stream-dwelling insects the Lambourn team has found very large variations in numbers from year to year, in an apparently natural situation. In other groups we found a very constant number from year to year. I think that Dr. CRICHTON'S results reflect very well the general picture in many, if not most, species of Trichoptera. There are some species which are very variable, and the hydroptilids seem to be notable in this respect. The group of limnephilids which have a long flight period seem, from my studies of larvae, to have very constant numbers from year to year. CRICHTON: There are of course fluctuations but it is striking how often at a given site the total catch of a species in one year may be, for example, perhaps two individuals, the next year one, and the following year three. HANSELL: What about the correlation between day and night flying with the size of the eyes or presence of ocelli? CRICHTON: There seems to be no correlation between the presence of ocelli and the time of day when caddis-flies are active. BADCOCK: I am very impressed that there is such close agreement between Dr. CRICHTON'S computer maps for hydropsychid imagines in light traps and my own maps based mainly on larval distributiori. This is seen clearly as regards the distribution of the more abundant species, e.g. the occurrence of Hydropsyche angustipennis predominantly in the midlands and south of Britain. There is an explanation of one difference, this is H. contubernalis in the Lake District. I have one habitat in the River Derwent where this species is extremely abundant but not near to light traps. It does occur in the Lake District but only in particularly large rivers. There does not happen to be a light trap in the right place. CRICHTON: There was in fact no light trap in the Lake District but H. contubernalis did not occur in the traps in Northern England. It was abundant in traps near rivers in the Midlands and Southern England as shown in the computer map. HIGLER: I was surprised to see that the numbers of Limnephilus affinis you showed were so low. In our country it is the commonest caddis, I think. Is this not the case in England? CRICHTON: This is a question of location of the traps. The larvae are tolerant of brackish conditions. Some of the traps in suitable coastal habitats did produce large numbers. HIGLER: We found high numbers in the night by light traps, but it is also one of the species that flies at daytime too, like more of the Limnephilids. HILEY: It is certainly a species of salt marshes, but also occurs in freshwater. I have analysed only one of the waters in which I found it, an inland pool where the salt concentration was rather high, but not sufficient to be noticed by taste. Is there perhaps a lot of salt still in the soils in your country? HIGLER: I do not consider this as a factor of importance, but most of our waters are more or less heavily polluted, so the conditions may be compared with brackish waters, I think. Perhaps this is the solution. This species is really not only found in the more brackish waters; it is not a river species, it inhabits ponds and even ephemeral waters.

158 Proc. of the First Int. Symp. on Trichoptera, 1974. Junk, The Hague

Pollution and caddis-fly fauna

ANKER NIELSEN

In a couple of decades the brooks and small rivers in the Danish peninsula Jylland (Jutland) have become among the most polluted in the World. This is chiefly, in many cases exclusively, due to the erection of about 800 fish-farms in this small part of Europe. Even at quite small rivers, scarcely rivers at all, several fish-farms may be found at a mutual distance of some few kilometres. In these farms the northwest American Salmo gairdneri (rainbow or steelhead) is kept in crowding cultures. Enormous amounts, in some fish-farms more than ten tons a day, of minced marine coarse fish are used as fodder. The greater part of this goes downstream as fish excrements and fish urine, a very heavy load of organic matter. I have witnessed the destruction of the once exceptionally rich fauna. Among the first caddis-flies to disappear are Oligopleetrum maeulatum FOURCR. and Eeclisopteryx guttulata PICf'. The latter pupates in gravel beneath stones and there• fore of course is very sensitive to deposition of silt. Hydropsyehe, especially H. angustipennis CURT., is favoured by some degree of pollution. I have in one year seen a very dense population (up to 10 000/m2) of Oligopleetrum maeulatum succeeded by a dense population of Hydropsyehe angus• tipennis. By increasing pollution Hydropsyehe disappears, and the fauna becomes dominated by blackfly larvae of the species groups Simulium equinum and S. ornatum. They feed on bacteria, and they are very tolerant to oxygen deficiency. The caddis-fly species disappear, one after the other. Surprisingly enough Rhyaeophila is among the more resistant. In the small, heavily polluted river Lindenborg Aa larvae of some of the hardiest species, Anabolia nervosa LEACH, Halesus digitatus SCHR. and Potamophylax latipennis CURT., may be found in small numbers. However, all the pupae found were dead, choked _by silt, so the presence of larvae must be due to immigration of adults from nearby, smaller and not so polluted streams. A number of caddis-flies - as well as other rheophile insects - which in Denmark were found only in Jutland probably now are extinct in this country, due to the pollution described above. To sum up: the caddis-fly fauna is a very good indicator of the health of running water. The slightest change in this fauna is a warning that something is wrong. I am aware that also my colleague KUMANSKI has drawn attention to the caddis-fly fauna as an indicator of pollution.

159 And may this be a warning to other countries against uncontrolled erection of fish farms.

Discussion

HILEY: First of all, is it more efficient to feed trout in your country on ground meat? In Great Britain we don't feed them on wet meat; we use dry pellets and these give less dissolved organic material to the water. They do, after going through the trout, give rise to some small pollution in the form of dissolved minerals and this is manifested by growths of algae. In Great Britain anyway Potamopyrgus jenkinsi is a fresh as well as brackish water snail, and it has a rather strange distribution from year to year. It can occur in very large numbers in one place one year, then in very low numbers the next year. This has been charted in many areas of Britain. I do not know what happens on the continent. Have you tested the oxygen concentration of the streams you say are polluted? Do you think there is strong evidence for pollution being the trout food or can it be pollution from other sources as well? NIELSEN: I know that the fish farmers claim that when they use dry food there is no pollution at all, and that the pollution is due only to spillage of food. This is not the case. Food spillage plays only a minor role in the pollution; the fish-farmers of course avoid spillage as far as possible, because food is expensive. The pollution is chiefly due to the excrements and the excretions from the fishes, and that cannot be prevented. I have made very many oxygen measurements in the streams, though not after the erection of fish farms, but I know that others have, and have found a very low oxygen content indeed below the fish farms. I have known the streams before and after the erection of fish farms. In the vast majority of the streams things - apart from the fish-farms - have been equal before, or even better, since there is not kept so much cattle in Denmark today as formerly. So I cannot see any other cause to the pollution than just the fish-farms. MORETTI: Nous avons en train une etude sur la pollution du fleuve Tibre en Italie, et nous avons pu voir que Hydropsyehe dissimulata, Hydroptila angulata et Eenomus tenellus sont les especes les plus resist antes aux pollutions. NIELSEN: As I mentioned, Hydropsyehe are rather tolerant to pollution, and even to some degree favoured by pollution. About Eenomus tenellus I don't know anything because in Denmark it is chiefly found in lakes. BOURNAUD: Do you think that the ecology of Trichoptera is so advanced that we can detect small pollution by examining populations of Trichoptera? NIELSEN: When the fish farms were erected, the water for some time seemed clean enough; the first thing I noticed was that Oligopleetrum maculatum disappeared from places where it formerly lived in numbers up to 10000/m2. And though the water still seemed clean enough, the Trichoptera fauna changed to these enormous populations of Hydropsyehe angustipennis. Even the least alteration of the Trichopterous fauna is a warning. As a result of this pollution of streams, I think that a few Trichoptera which from Denmark was known only from this region now must be considered extinct in Denmark. NEBOISS: In one of the large rivers in Victoria we have come across a somewhat similar problem where disappearence of some caddis-fly species is evident. The only difference here is that we do not know what exactly existed before the pollution reached its present level. The river starts from the mountains, flows through extensively cultivated agricultural districts and finally through densely populated industrial zone with paper mills, electric power station and other industries. The survey showed definite reduction in species content from 60 species in 12 families in upstream localities to 5 or 6 species in 2 families downstream of the industrial zone.

160 Tests on a similar stream where there is no industrial zone showed less changes in species numbers. The evaluation of data is in progress. NIELSEN: Of course there are other causes to pollution than fish farming, but it is the chief cause in the peninsula Jutland. HILEY: Have you, apart from the obvious one of eliminating trout farming, ideas as to how to lessen the effects of the trout farms? NIELSEN: Just abandon trout farms. In my country I always object to the word trout farming, for trout is not Salrna gairdneri, trout is Salrna trutta. LHONORE: Actuellement, dans mes recherches histophysiologiques sur les Trichopteres, j'ai la possibilite de detecter tous les elements de la classification periodique (11 partir du Beryllium) qui deposent sous la forme de concretions dans les divers organes de ces Insectes. Je recherche des Trichopteres vivant dans les eaux saines a titre de reference mais aussi des Trichopteres vivants, proven ant d'eaux polluees par des composes inorganiques (ex: Ie mercure comme vient d'en parler Ie Dr. NIELSEN). La recherche des concretions dans les tissus de ces animaux et leur analyse permettraient peut-etre de connaitre les seuils de resistance aux divers agents poilu ants mineraux deverses dans les eaux. Cette etude est possible grace 11 I'histochimie, la spectro• graphie des rayons X (Microsonde de CASTAING) et 11 I'analyseur par emission ionique secon• daire. NIELSEN: Not quite small amounts of mercury have been found in the outflow from fish farms in Denmark. It would also be possible to make observations on the contents for instance of mercury in caddis-fly larvae; it has not been made. But I know that in some Danish Fjords there have been made analyses of mercury contents in Polychaeta, and it has been compared with mercury contents in old material in our museum in Copenhagen; there is also mercury in the old specimens, there is a background of mercury, but it has been shown that in recent specimens of Polychaets there is much more mercury.

161 Proc. of the First Int. Symp. on Trichoptera, 1974. Junk, The Hague

Revision of some opinions expressed in my 1942 paper

ANKER NIELSEN

Many insects, mayflies, stone-flies and caddis-flies inhabiting fast flowing water in Holsten and the Danish peninsula Jylland (Jutland), and especially those living in spring brooks with a low summer temperature, have been considered as relicts from a period with a colder climate, sometimes even as Glacial relicts. In 1942 I published a paper, based on studies of streams in the province Himmerland in northern 'Jutland, a somewhat karstic area rich in large springs. In this paper (pp. 622-625) I reduced the number of relicts considerably, and pointed out that some of the spring inhabiting species might be relicts from the Atlantic Period in the Post Glacial, in which the winter is believed to have been warmer than nowadays. If so, they had survived in the springs due to the relatively high winter temperature of the latter. I now realize that I did not reduce the number nearly enough. Actually the number of relicts is very small. It is certain that Apatania muliebris McL. is a Glacial relict. During the last glaciation the southwestern part of Jutland was not covered by ice. It was a glacial delta, above which some low hills rose. Apatania muliebris probably lived on these hills. The climatic conditions must have been similar to those now prevailing in Alpine and Norwegian high mountains, where the species is widespread. It is probable that Parachiona picicornis PICT. is a relict from the Late Glacial, when the ice had retreated to the Scandinavian mountains, but the forest not yet invaded Denmark. And it is possible, just possible, that Odontoceium albicorne Scop. is a relict from the Atlantic Period. However, the trichopterus fauna in the springs is chiefly composed of species which are resistant to adverse life conditions. To take an example, the two Danish species of Silo, nigricornis PICT. and pallipes F. In Himmerland nigricornis lives in spring brooks, pallipes in streams which have a high summer temperature. Both species occur also on the Danish main island, Sjaelland. Here nigricornis is found in brooks which attain a high summer temperature, and which in dry periods almost may stop flowing. It is found in these brooks because it is resistant to an occasionally very slow current, in the spring brooks in Himmerland because it is resistant to a low summer temperature. The two Danish species of Rhyacophila, fasciata HAGEN and nubila ZETT., are an exception to the rule that it is the resistant species which lives in springs. In Himmerland fasciata is characteristic of spring brooks, nubila of small rivers with a high summer temperature. But in Norway nubila is abundant in torrents coming

163 directly from glaciers and very cold in summer indeed. R. nubila is resistant both to high and to very low summer temperatures, and between these extremes it cannot compete with fasciata. In Himmerland the occurrence of Agapetus fuscipes CURT. is restricted to streams in which the temperature varies between 4-12 0c. The lowest temperatures occur on winter days with much snow melting, a frequent phenomenon in Denmark. On the island Sjaelland Agapetus fuscipes is characteristic of small forest brooks which in winter may freeze completely, leaving only a small trickle of water at the bottom, and which in summer may dry almost entirely up. Agapetus fuscipes is resistant to these adverse conditions, as it in the springs of Himmerland is resistant to the low summer temperature. In the forest brooks on Sjaelland Agapetus fuscipes has an early and short flying season as described by THIENEMANN in springs in Holsten. In Himmerland a special spring form has evolved. Here there is no definite flying season, but imagines emerge during the whole year, even in the middle of winter. The females which emerge in winter probably do not propagate, but the populations are able to bear this loss. By the way, in the same localities Wormaldia occipitalis PICT. may emerge in the middle of winter. There were three populations of Apatania muliebris McL. in Denmark. In two of these -the species lived only in the uppermost about 50 meters of the spring brooks, where the temperature never reached 8 centigrades. In each spring the entire population probably amounted to a couple of thousands. By population is meant the number of specimens which reached maturity. In the third locality, Lille Blaakilde, the inhabited area was even more restricted, and the whole population probably only some few hundreds. A young biologist wanted to improve my studies on Apatania muliebris, and to this end he needed a material great enough for a thorough statistical analysis. He therefore made intensive and regular collections, and in this way he eradicated the populations in Rold Kilde and Lille Blakilde. In many years I have at least sought in vain for larvae and pupae in these two localities. The two endemic subspecies, intermedia NIELSEN and nielseni SCHMID, must now be considered as extinct. In the third Danish locality the population is still intact.

Reference

NIELSEN, ANKER 1942. Ueber die Entwicklung und Biologie der Trichopteren mit besondere Beriicksichtigung der Quelltrichopteren Himmerlands. Arch. Hydrobiol. Suppl. 17: 255-631.

Discussion

BOTOSANEANU: I was very impressed to hear what Dr. NIELSEN says about the eradication of some Danish populations of Apatania muliebris. It is indeed a very sad thing. This reminds me to ask you for information about Trichoptera species or populations which have become extinct in some areas or which are on the verge or in danger of extinction, on Trichoptera habitats in

164 danger of being destroyed or already destroyed. I know, there are many such examples. Please send me all available information as soon as possible. I intend to publish one or several papers in periodicals devoted to the protection of nature. All contributions will be warmly acknow• ledged. This seems to be our most urgent task (when I say 'our', I mean Trichopterists). Please transmit this to all colleagues not attending this symposium. Thank you very much. MALICKY: Near one of my streams here in Lunz (the Schreierbach) Wormaldia copiosa and Philopotamus ludificatus are present in the adult stage throughout the year, even in the winter when one or two meters of snow covers the soil, but this is only true in the immediate spring region. In the rest of this stream, under the same temperature conditions (this is a karstic stream with temperatures, independent from the season, between 5.5 and 7.0°C) the adults have a long activity period, but not in winter. I don't think that there is a distinct genetically fixed spring population, but I rather suppose that the development of these species in the spring is affected by any environmental factor. I have no idea which one, but it is not temperature. HILEY: Can I say something about Apatania muliebris? In Britain there are many populations of A. muliebris of various subspecies which I have not determined yet, and the larvae are very variable. Most of these populations are very limited, and in some springs I studied there were perhaps 30 or 50 individual larvae, inhabiting a very short zone about 10-25 m from the source. Obviously any study that involves larval removal is going to destroy the population very quickly. I would like to mention Dr. ELLIorr's paper which describes his study of A. muliebris in a stream near Windermere. In this study he was very careful to preserve the population; he always measures larvae alive, in the field, and returned them immediately to the stream. In this way he did not damage the population at all. NIELSEN: When I studied muliebris, I of course also made collections at regular intervals, but I never collected more than 24 specimens, and I tried to make the collections as representative as possible. It is of course a bit subjective, but on the other hand, I carried the investigations out through 10 years, so I believe it to be a rather reliable analysis.

165 Proc. of the First Int. Symp. on Trichoptera, 1974. Junk, The Hague

Changes in the caddis-fly fauna of Lake Erie, Ohio, and of the Rock River, Dlinois, over a fifty year period of environmental deterioration

VINCENT H. RESH

This report is the first of a series that will deal with techniques for the development of water quality tolerances of species of North American caddis-flies. In this study, the changes in the caddis-fly fauna of two regions in the midwestern United States that have undergone drastic environmental deterioration over the past fifty years, are used in analyzing the potential of the caddis-fly genus Athripsodes (Trichoptera: Leptoceridae) as indicators of water quality. The faunal changes in the caddis-fly populations are documented by a re-examination of the collecting sites, records and specimens of past studies of these areas that are currently maintained in university and museum collections. The information compiled in this study emphasizes the value of using museum and university collections in environmental assessments. In addition, the role and importance of the systematic biologist in environmental llssessment and the development of the indicator organism concept is further delineated. The life histories of North American Athripsodes are of two basic types. The first group consists of the detritivores, which are characterized by a univoltine life cycle, five discrete instars and a recognizable single cohort from the time of larval hatching through pupal emergence. There is a short period of adult flight in the spring. The second group consists of the sponge feeders, a morphologically distinct assemblage of species that share their unusual feeding habit with spongilla flies (Neuroptera: Sisyridae) and a single species of chironomid, Xenochironomus xenolabris (KIEFFER). The life cycles of the sponge-feeding species of Athripsodes are mor~complex than those of the detritus-feeding species. A common species, Athripsodes angustus has two cohorts occurring in the population, one of which feeds on freshwater sponge through all five larval instars, overwinters as a prepupa and then has a brief adult emergence period early in the spring. The larvae of the next generation of this cohort result from eggs laid during this flight period. The second cohort develops from eggs laid during a second adult emergence period that occurs in late summer. However, the larvae of this cohort will only reach the third or possibly fourth instar before the sponge begins to break down in the autumn and gemmulation for overwintering occurs. These overwintering larvae must then switch to detritus feeding, returning to the sponge feeding habit only when the 167 sponge reappears in the spring. The fourth and fifth instar larvae of this second cohort are significantly smaller (using instar-specific head width measurements) than the corresponding instars of the cohort that fed entirely on freshwater sponge and then overwinter as a prepupa. The emergence of this second cohort finally occurs m late summer. In addition to diverse life histories, the genus Athripsodes is also characterized by species with a wide range of water quality tolerances. While the tolerances do not fall into the groups deliniated by feeding habits, it should be remembered that the survival of the sponge feeding species necessitates inhabiting an area of sufficiently high water quality to insure survival of the sponge. In examining faunal changes in Athripsodes distribution, the early collections and detailed field notes of R. E. RICHARDSON deposited at the Illinois Natural History Survey provided baseline information in developing water quality criteria for Ath• ripsodes caddis-flies. From 1924-1927, RICHARDSON made extensive collections with detailed locality descriptions along the Illinois portion of the Rock River. Additional collections were made by H. H. Ross from several locations along the Rock River in the late 1930's and the early 1940's. From examinations of the collections and locality designations of these collectors, localities where Athripsodes menteius were found in earlier studies could be determined. This species was the dominant leptocerid caddis-fly in the collections of RICHARDSON. In 1971 and 1972, the Rock River was examined at four sites where A. mente ius was abundant in the earlier collections. Neither immature nor adult specimens of this species was found, although large numbers of Athripsodes transversus, especially rare in RICHARDSON'S collections, were collected. It is interesting to note that in his analysis of Illinois streams, SMITH (1971) reported that the Rock River flows through areas of urbaniza• tion and industrialization with a resulting deterioration in water quality. In a second example, changes in the distribution of caddis-fly species in Lake Erie were examined. The presence of species endemic to Lake Erie in collections taken during the early 1930's and deposited in the Entomology Collection at the Ohio State University provided the impetus for the investigation into the current status of these Athripsodes populations. MARSHALL (1939) reported the results of the exten• sive light trap collections from Lake Erie near Put-in-Bay, Ohio, in 1937. Of the nine species of Athripsodes collected, the following were listed as common: A. angustus, A. cancellatus, A. erullus, A. resurgens, A. saccus and A. tarsipunctatus. Specimens of A. erraticus, also collected from Put-in-Bay and currently in the Ohio State collec• tions, indicate that this species was also quite common in Lake Erie in the 1930's. In a more recent study, HORWATH (1964) reported the results of extensive blacklighting from the same location as MARSHALL'S earlier work. Only 4 of the original 9 species were present: A. an gustus, A. cancellatus, A. resurgens, and A. tarsipunctatus. The total numbers collected and the number of collections in which each species appeared were also reduced. The entire area was again examined intensively by me in 1972 and the only species present were those found in HORWATH'S 1964 study.

168 It appears that the extirpation from Lake Erie of four of the common Athripsodes species - A. erullus, A. erraticus, A. saccus and A. submacula - is now complete. The more tolerant species of Athripsodes remain, but in greatly reduced numbers. In presenting this information about the diverse ecological types and the change~ in distribution over the past fifty years in the Rock River and Lake Erie, it is important to emphasize that if identifications of the species involved had been made at the generic level no differences or changes in species composition would have been noted. Recently, there has been a tendency for biologists involved in environmental studies to concentrate on generic level identification and then to attempt to arbitrar• ily assign this taxonomic level a pollution-tolerant or pollution-intolerant designa• tion. The futility of attempting to develop water quality criteria using indicator organ• isms that have been identified only to the generic-level is summarized in RESH & UNZICKER (1975) and in Table 1. In this tabulation, the numbers of genera of aquatic macroinvertebrates for which water quality tolerances to decomposable organic wastes have been established for more than a single species (in that genus) are listed according to the arbitrary assignment of the individual species water quality tolerances. This information is compiled from Table 7 of the Macroinvertebrate section of WEBER (1973). In that review, the index species are classified according to three arbitrary categories: tolerant, facultative, or intolerant. Of the 87 genera for which water quality tolerances have been established for more than a single species, the component species fell into different tolerance categories in 60 of the genera examined. The largest group of genera in Table 1 belong to the category in which some species were judged tolerant to pollution, while other species in the same genus were judged either intolerant or facultatively tolerant to pollution. This table is a summary of our current state of knowledge concerning indicator organisms and perhaps better than any other example emphasizes the need for species-level identifications in ascertaining water quality tolerances. Above all else, it signifies the questionable value of the generic-level taxonomic unit as a water-quality indicator.

Table 1. Numbers of benthic macroinvertebrate gen• era for which water quality tolerances have been es- . tablished for more than a single species. (Data from Table 7, Macroinvertebrates, in WEBER 1973). Ab• breviations: T, tolerant frequently associated with gross organic contamination; F, facultative, frequently associated with moderate levels of organic contamina• tion; I, intolerant, not found associated with even moderate levels of organic contaminants.

T T/F F T/F/I F/I

5 10 5 28 22 17

169 Unfortunately, in most groups of aquatic insects, identification of the immature stages cannot currently be made below the generic level. This becomes a critical problem as the immature stage in the life cycle of an aquatic insect is the one most commonly encountered by hydrobiologists. There is a critical need for increased support of investigations to develop identifi• cation keys and describe basic life histories of these aquatic macro-invertebrates. It is also important that ecological agencies begin to support the programs of taxonomists in collecting data and producing the publications which are sorely needed. The time and effort spent on identifying specimens to genus, in developing endless and meaningless generic-level faunal lists in the preparation of environmental assessments should be shifted to associating immature and adult aquatic insects, and developing appropriate identification keys. As a consequence of this effort, the species lists prepared in the future will not merely be a taxonomic exercise but instead, a valuable tool in the biological assessment of water quality. References

HORWATH, A. B. 1964. Insects of Gibralter Islands in relation to their habitats. Unpublished M.S. thesis. Ohio State University, 61 pp. MARSHALL, A. C. 1,)3,). A qualitative and quantitative study of the Trichoptera of Western Lake Eric (as indicated by light trap material). Ann. Entomol. Soc. Amer. 31: 665-688. RESH, V. H. & J. D. UNZICKER. 1975. Water quality monitoring and aquatic organisms: the importance of specific versus generic identifications. 1. Water Poll. Contr. Feder. 47: 9-19. SMITH, P. W. 1971. Illinois streams: A classification based on their fishes and an analysis of factors responsible for disappearance of native species. III. Nat. His!. Surv. BioI. Notes no. 76. 14pp. WEBER, C. 1. 1973. Biological field and laboratory methods for measuring the quality of surface waters and effluents. National Research Center, Environmental Protection Agency, Cincin• nati, Ohio. Environmental Monitoring Series. Program Element IBA027. EPA-670/4-73- 001. 138 pp. Discussion

FLINT: As you say, I think this points up the fact that most diversity indexes are very poor. At recent national meetings, I have seen indices with a few families listed, many genera, and some species, yet each is counted as one, and an index computed. Obviously, such indices are very, very poor and extremely misleading. RESH: An expanded version of the material presented in this talk that deals with the importance of specific-level versus generic-level identifications and their use in water quality studies will appear early in 1975. Following the recommendations of several of the participants at this symposium, we have decided not to publish this paper in an entomological journal, but rather in a periodical that is geared more towards the administrators and the engineers involved in water quality monitoring programs, the Journal of the Water Pollution Control Federation. This article emphasizes the need for supporting research into developing the specific-level larval identification keys that are critical in environmental studies.

170 Proc. of the First Int. Symp. on Trichoptera, 1974. Junk, The Hague

Morphologische Komponenten bei der Osmoregulation von Trichopterenlarven

WILFRIED WICHARD

Abstract Morphological components in the osmoregulation of caddis-fly larvae. Caddis-fly larvae possess anal papillae or abdominal chloride epithelia, which are involved in osmotic hyperregulation by the absorption of salt from the surrounding water in order to compensate for the electrolytes lost by the fluid excretion through the excretory organs. Finestructurally the epithelia of anal papillae and the chloride epithelia display the features of transporting epithelia, and are characterized by deep infoldings of the basal and apical plasma membranes which are associated with abundant mitochondria.

Die Larven von Kocherfliegen leben in der Regel im SiiBwasser. Daneben werden sie vereinzelt im Brackwasser beobachtet, etwa im finnischen Meerbusen (SILTALA, 1906) oder im marinen Gezeitenbereich (MALICKY, 1974). Bezogen auf die hohe Ionenkonzentration in der Hamolymphe (SUTCLIFFE, 1962b; BEAUJOT, NAOUMOFF & JEUNIAUX, 1970) befinden sich die Larven des SiiB- und Brackwassers im hypotonischen Milieu. Die wenigen marinen Trichopterenarten leben dagegen im hypertonischen Milieu. Urn dem hohen osmotischen Druck im Meerwasser zu widerstehen, sind diese Larven offenbar zur hypoosmotischen Regulation befahigt (LEADER, 1971, 1972). Die Regulation der im SiiB- und Brackwasser lebenden Larven ist im aHgemeinen hyperosmotisch. Sie tolerieren nur geringe Salzkon• zentrationen und sind bei euryhalinen Arten fakultativ auch zur osmotischen Hypo• regulation befahigt (HAAGE, 1968, 1969; SUTCLIFFE 1960, 1961a, 1961b, 1962a). Die Osmoregulation ist nun eine notwendige Begleiterscheinung nicht nur der Exkretion sondern vor aHem der Respiration. Die Larven der Kocherfliegen haben ein geschlossenes Tracheensystem. Die Sauerstoffdiffussion findet auf der Korperoberfliiche als Hauttracheenatmung statt. Diese respiratorische Oberfliiche ist bei vielen Trichopterenarten dariiberhinaus durch eine hohe Anzahl von Tracheen• kiemen vergroBert. Die Diffussion des Sauerstoffs steht dabei im Zusammenhang mit der Diffussion von Wasser entsprechend dem ionalen Konzentrationsgefiille zwischen dem AuBenmedium und der Hiimolymphe; denn ein permeables Integument fUr Sauerstoff ist aufgrund der MoiekiilgroBe meist auch permeabel fur Wasser.

171 Bei den Larven von Kocherfiiegen, die als Bewohner salzarmer Gewasser im hypotonischen Milieu leben, diffundiert durch die Korperoberfiache demzufolge Sauerstoff und Wasser. Der Sauerstoff sammelt sich in den subcuticularen Tracheolen des Tracheensystems und das Wasser im hypertonischen Hamolymphraum. Dieser osmotische Wasserstrom wird unter Beteiligung von Elek• trolyten wieder aus dem Hamolymphraum ausgetrieben, indem die Elektrolyte zunachst transepithelial in das Lumen der Malpighischen GefiiBe gepumpt werden. Dem osmotischen Sog folgend wird der WasseriiberschuB ebenfalls in das Lumen der Malpighischen GefiiBe gebracht und anschlieBend nach teilweiser, rektaler Riickresorption von Elektrolyten durch den Anus als hypotonischer Ham ausges• chieden (vgl. BONE & KOCH, 1942) (Abb. 1a).

a b

Abb. 1. Schematische Darstellung der wichtigsten Elektrolyt- und Wasserverschiebungen im Rahmen der osmotischen Hyperregulation (Wasser: gespaltene Pfeile; Elektrolyte: unter• brochene Pfeile) bei Larven mit ionentransportierenden Analpapillen (a) und ionentranspor• tierenden Chloridepithelien (b).

Bei der, stark vereinfacht dargestellten, osmotischen Hyperregulation tritt ein Elektrolytverlust ein, der durch eine aktive Ionenaufnahme aus dem wassrigen Milieu ausgeglichen werden muB. Als Orte fiir diese notwendige Salzaufnahme kommen bei den Larven der Kocherfiiegen nach elektronenmikroskopischen und

172 histochemischen Befunden Analpapillen und abdominale Chloridepithelien in Frage (Abb. la,b) (NOSKE & WICHARD, 1971, 1972; WICHARD & KOMNICK, 1973). Die Analpapillen wurden bislang oft als 'Blutkiemen' bezeichnet, da bei ihnen statt eines lonentra9sportes eine respiratorische Aktivitiit zwischen dem AuBenmedium und der Hiimolymphe vermutet wurde. Die abdominalen Chloridepithelien mancher Kocherfliegenlarven wurden nach KRAWANY (1935) als 'Kiemenplatten' oder 'At• mungsfelder' beschrieben, wei! diese Felder nach Behandlung mit Silbernitrat starke argyrophile Reaktionen zeigen, die KRAWANY (1935) mit der Reduktion von Sil• berionen durch die respiratorische Aktivitiit dieser Felder erkliirte. Die Feinstruktur der abdominalen Chloridepithelien (Abb. 2a) und der Epithelien von Analpapillen (Abb. 2b) weist typische Merkmale des aktiven Transportepithels auf. Die einschichtigen Epithelien werden auBen von der Cuticula und zum Hiimolymphraum von einer Basalmembran begrenzt. Sie heben sich von der umgebenden Hypodermis deutlich durch ihre unterschiedlich differenzierten Zellen abo Auffallend ist eine hohe Anzahl von Mitochondrien, die mit dicht gepackten Cristae auf eine groBe Stoffwechselaktivitiit der Zellen hinweisen. Die apikale Oberfliiche dieser Epithelzellen wird durch tiefe, fast parallel verlaufende Einfal• tungen der apikalen Membran erheblich vergroBert. Zur Cuticula hin nehmen die Einfaltungen stellenweise die Form von Mikrovilli an. Oft dringen die Mitochondrien in den Bereich der apikalen Einfaltungen und bilden einen engen Kontakt zu den apikalen Zellmembranen. Die basale Zellmembran bildet durch Faltenwiirfe und Verzahnungen ein basales Labyrinth, das aus tiefen und oft ampullenartig erweiter• ten Invaginationen besteht und sich bis in den apikalen Bereich erstreckt. Diese OberfliichenvergroBerung korreliert mit cristaereichen Mitochondrien zu einem ver• mutlich funktionellen Komplex. Ebenfalls unregelmiiBig und gewunden verlaufen die lateralen Zellmembranen und erfiillen im basalen Zellbereich moglicherweise diesel• ben Funktionen wie die Invaginationen der basalen Zellmembran. 1m apikalen Bereich sind die Zellen iiber Desmosomen und septierte Desmosomen verbunden. Neben diesen spezifischen Struckturen enthalten die Epithelzellen unregelmiiBig verteilt Tracheolen, die ebenso in der umgebenden Hypodermis gefunden werden und sehr wahrscheinlich der allgemeinen Hauttracheenatmung auf der Korperoberfliiche dienen; die Respiration erfolgt hierbei nicht transepithelial zwis• chen dem AuBenmedium und der Hiimolymphe, sondern iiber das Tracheensystem. Fiir eine funktionelle Interpretation der Analpapillen als 'Blutkiemen' oder der Chloridepithelien als 'Atmungsfelder' gibt es keinerlei cytologische Anhaltspunkte. Eine weitere Stiitze erhiilt die funktionelle Interpretation der Analpapillen und Chloridepithelien als ionentransportierende Epithelien durch den histochemischen Nachweis von Chloriden (NOSKE & WICHARD, 1971; WICHARD & KOMNICK, 1973). Bei der Behandlung mit Silbernitrat werden die Chloridepithelien bei Auflicht weiB sichtbar. Die Farbe wechselt ins Violett oder Dunkelbraun, wenn sie dem hellen Tageslicht ausgesetzt wird. Nach Behandlung mit Salpetersiiure selbst bei Benutzung einer 10 N Konzentration losen die Priizipitate sich nicht auf; eine entsprechende Behandlung mit Ammoniumcarbonat lost sie dagegen vollstiindig auf. Diese

173 Abb. 2. Schematische Darstellung des ionentransportierenden Chloridepithels (a) und (b) des ionentransportierenden Epithels von Analpapillen. Das Chloridepithel (a, rechts) wird von einer Grenzstruktur mit reduzierter Cuticula (a, mittel gegen die umgebende Hypodermis (a, links) abgesetzt.

chemisch-analytischen Ergebnisse zeigen unzweifelhaft, daB die Prazipitate aus Sil• berchlorid bestehen, was auBerdem durch Feinbereichselektronenbeugung direkt bewiesen ist. Die Argyrophilie, die KRAWANY (1935) beschreibt, resultiert daher aus der sekundaren, photochemischen Reduktion von Silber, wahrend die entscheidende primare Reaktion in der Silberchloridfiillung besteht und auf die Akkumulation von Chloriden hinweist. Elektronenoptisch werden dichte Priizipitate nach Fixation mit Os04/Silberlactat in der Cuticula oberhalb der Chloridepithelien nachgewiesen; sie Jiegen hauptsachlich in der Epicuticula und sind sehr gering in der Pro cuticula verstreut. 1m Epithel der Analpapillen lassen sich die Silberchloridprazipitate als granulare Reaktionsformen gehauft im mittleren Bereich der Zellen zwischen dem apikalen Faltensaum und dem basalen Labyrinth nachweisen. Die Analpapillen bei den Larven von Kocherfliegen liegen praeanal; sie werden durch Hamolymphdruck augestiilpt und mit Muskeln wieder eingezogen. Die Larven besitzen gruppenspezifisch drei bis sechs Analpapillen. Hiervon abweichend glaubt STADLER (1942) bei Lasiocephala basalis verzweigte Blutkiemen zu beobachten. Als

174 Blutkiemen hat LUBBEN (1907) die Analpapillen zusammenfassend dargestellt. Eine erste sorgfiiltige Histologie der Blutkiemen beschreibt HALLER (1948). THIENEMANN (1903) beobachtet im zentralen Hamolymphraum von Analpapillen der Glos• sosomatiden langsverlaufende Tracheen, die sich der epithelialen Wand der Anal• papillen nahern, und vermutet eine respiratorische Funktion. Elektronenmikros• kopische Befunde bestatigen diese Annahme (NUSKE & WICHARD, 1972) und deuten daraufhin, daB neben der dominierenden Funktion des Ionentransportes auch im Epithel der Analpapillen eine Hauttracheenatmung stattfinden kann. Die Chloridepithelien, die als 'Atmungsfelder' oder 'Kiemenplatten' zunachst von KRAWANY (1935) beschrieben wurden, befinden sich auf den abdominalen Segmen• ten als runde und ovale Felder und sind nach der histochemischen Behandlung einwandfrei zu erkennen. Die Verteilung der Chloridepithelien auf dem Abdomen ist art- und gruppenspezifisch. Unter den europaischen Trichopteren besitzen die Limnephiliden, Goeriden und Hydroptiliden Chloridepithelien; aber auch die Molanniden, nachdem sich durch histochemische Untersuchung gezeigt hat, daB die von BETTEN (1902) beschriebenen Drusen ebenfalls Chloridepithelien sind. Die von KRAWANY (1935) als 'Kiemenstreifen' oder als 'diffuse Kiemenfelder' beschriebenen Strukturen erweisen sich als perlschnurartige Aneinanderreihungen oder als lockere Anhaufungen von punktfOrmigen Silberchloridprazipitaten. Feinstrukturell konnen sie nicht in Zusammenhang mit ionentransportierenden Zellen und Epithelien geb• racht werden, sondern mussen wahrscheinlich als diffuse Akkumulationen aufgrund der chemischen und strukturellen Beschaffenheit der Cuticula verstanden werden.

Literatur

BEAUJOT, J., NAOUMOFF, M. & JEUNIAUX, CH. 1970. Les cations inorganiques de I'hemolymphe larvaire des Insectes Trichopteres. Arch. internal. Physiol. Biochem. 78: 11-118. BETTEN, C. 1902. The larva of the Caddis-Fly Molanna cinerea HAGEN. J. N.Y. Ent. Soc. 10: 147-154. BONE, G. & KOCH, H. J. 1942. Le role des tubes de Malpighi et du rectum dans la regulation ionique chez les insectes. Ann. Soc. roy. Zool. Be\g. 73: 73-87. HAAGE, P. 1968. On the salinity tolerance of eggs and young larvae of Phryganea grand is LINNE (Trichoptera). Hydrobiologia 32: 257-270. --. 1969. Salinity preference and tolerance of caddis larvae (Trichoptera). Opucs. Ent., Lund 34: 73-84. HALLER, P. H. 1948. Morphologische, biologische und histologische Beitrage zur Kenntnis der Metamorphose der Trichopteren (Hydropsyche). Mit. Schw. Ent. Ges. 21: 301-360. KRAWANY, H. 1935. Trichopterenstudien. X. Untersuchungen tiber die Atmungsorgane der Larven. Int. Revue d. ges. Hydro. u. Hydrogr. 32: 241-264. LEADER, S. P. 1971. Effect of temperature, salinity and dissolved oxygen concentration upon respiratory activity of the larva of Philanisus plebe ius (Trichoptera). J. Insect Physiol. 17: 1917-1924. LEADER, J. P. 1972. Osmoregulation in the larva of the marine caddisfly, Philanisus plebe ius (WALK.) (Trichoptera). J. Exp. BioI. 57: 821-838. LUBBEN, H. 1907. Uber die innere Metamorphose der Trichopteren. Zoo I. Jb. (Anat.), Jena 24: 71-128.

175 MALICKY, H. 1974. Eine im marinen Gezeitenbereich lebende europaische Trichopterenlarve. Arch. Hydrobiol. 73: 266-269. 'JOSKE, H. & WICHARD, W. 1971. Die Analpapillen der Kocherfliegenlarven. I. Feinstruktur und histochemischer Nachweis von Natrium und Chlorid bei Philopotamus montanus DONOV. Cytobiologie 4: 480-486 . ._- & --. 1972. Die Analpapillen der Kocherfliegenlarven. II. Feinstruktur des ionen• transportierenden und respiratorischen Epithels bei Glossosomatiden. Cytobiologie 6: 243- 249. SILTALA, A. 1. 1906. Zur Trichopterenfauna des Finnischen Meerbusens. Acta Soc. Fauna Fenn. 28: 1-21. STADLER, H. 1942. Blutkiemen bei einer kocherbauenden Trichopterenlarve (Lasiocephala basalis). Arch. Hydrobiol. 15: 250-252. SUTCLIFFE, D. W. 1960. Observations on the salinity tolerance and habits of a euryhaline caddis larva, Limnephilus affinis CURTIS (Trichoptera:Limnephilidae). Proc. R. Ent. Soc. Lond. (A) 35, 156-162. --. 1961a. Studies on salt and water balance in caddis larvae (Trichoptera). I. Osmotic and ionic regulation of body fluids in Limnephilus affinis CURTIS. 1. Exp. BioI. 38, 501-519. --. 1961b. Studies on salt and water balance in caddis larvae (Trichoptera). II. Osmotic and ionic regulation of body fluids in Limnephilus stigma CURTIS and Anabolia nervosa LEACH. 1. Exp. BioI. 38, 521-530. --. 1962a. Studies on salt and water balance in caddis larvae (Trichoptera). III. Drinking and excretion. 1. Exp. BioI. 39, 141-160. --. 1962b. The composition of haemolymph in aquatic insects. 1. Exp. BioI. 39, 325-343. WICHARD, W. & KOMNICK, H. 1973. Fine structure and function of the abdominal chloride epithelia in caddisfly larvae. Z. Zellforsch. 136, 579-590. THIENEMANN, A. 1903. Analkiemen bei den Larven von Glossosoma boltoni CURT. und einigen Hydropsychiden. Zoo I. Anz. 27, 125-129.

Discussion

MORETTI: Dans les larves qui sont a la fin de leur developpement, ou qui sont en passage des autres stades, est-ce qu'on trouve la situation decrite? WICHARD: ErwartungsgemaB besitzen Kocherfliegen in allen Larvenstadien Analpapillen oder Chloridepithelien. Die Anzahl der Analpapillen ist nach unseren Untersuchungen wahrend der larvalen Entwicklung konstant. Die Anzahl der abdominalen Chloridepithelien nimmt dagegen von Larvenstadium zu Larvenstadium bestandig zu; wobei das erste Larvenstadium moglicherweise keine hat. Bei den Stenophylacini und Chaetopterygini kommen im letzten Larvenstadium auf der abdominalen Ventralseite stets 6 ChloriOepithelien vor; eine Ausnahme macht die terrestrische Enoicyla, die keine Chloridepithelien besitzt. Die groBte Variabilitat in Anzahl und Verteilung weisen die Limnephilini auf. Es wird zur Zeit gepriift, ob bei diesem Tribus die Chloridepithelien als systematisches Merkmal herangezogen werden konnen. Ross: Is this the same mechanism that is used by mosquitos (Culicidae) and some other Diptera? WICHARD: Yes, the anal papillae. Unter den aquatischen Dipteren gibt es bei den Nematoceren-Larven Anaipapillen und bei einigen Cyclorhaphen (z.B. Eristalinus); ferner bei Kaferlarven (Elodidae; nicht aber bei den Elminthiden, die am Anus Tracheenkiemen besitzen) und bei Trichopterenlarven. Bei einer zweiten Gruppe aquatischer Insekten gibt es Chloridzel• len. Sie befinden sich nicht in einem epitheIialen Verband, sondern liegen verstreut im

176 Integument dieser Tiere: Ephemeroptera, Plecoptera und Hydrocorisae. Drittens werden bei aquatischen Insekten Chloridepithelien nachgewiesen; die abdominalen Chloridepithelien bei Trichopteren, die analen Chloridepithelien bei brachyceren Dipteren und die rektalen Chloridepithelien bei den Larven der Odonaten (Anisoptera und Zygoptera). Ross: Did you study the marine genus Philanisus? WICHARD: No. Diese ionenabsorbierenden Organe wiirden bei der osmotischen Hyperregula• tion der Meerestiere wegfallen oder eine exkretorische Funktion iibernehmen. Bei Stichlingen (Pisces), die auf den Kiemen Chloridzellen haben, arbeiten sie im marinen Bereich ex• kretorisch, im SiiBwasserbereich (hypotonisch) ionenabsorbierend. Die Funktion kann also, mindestens bei den Fischen, umgeschaltet werden. Feinstrukturell sind die Chloridzellen der Fische mit den Chloridzellen bei aquatischen Insekten in bestimmten Gruppen sehr identisch. ILLIES: I have to give a little historical remark. The name of KRAWANY was very often mentioned in the speech. He was the earlier pioneer of the Limnochemistry in freshwater insects. And Lunz is his place, here. He was the only man working on insects in the long history of the Lunz Biological Station before HANs MALICKY came up here and started his work. So we are on the historical ground of KRAWANY'S steps. VAILLANT: How is the situation under different salinities? WICHARD: We have studied the effect of different salinities of the chloride cells of mayfly larvae. The numbers of chloride cells correlate to the osmotic pressure in water. This interpretation also might be possible in Trichoptera. Morphological adaptations under different salinities and studies in saline lakes near Lake Neusiedl show that the area of chloride epithelia is greater if the salinity is smaller, and on the high salinity we find smaller area of chloride epithelia. These results suggest that the adaptation in chloride epithelia is correlated with the osmoregulatory situation.

177 Proc. of the First Int. Symp. on Trichoptera, 1974. Junk, The Hague

A progress report on studies on the functional morphology of the genitalia in three new species of Cheumatopsyche (Hydropsychidae)l

BERNHARD STATZNER

A description is given of the morphology of male and female genitalia of three species from Central Africa of the genus Cheumatopsyche WALLENGREN, Ch. explicanda sp.n., Ch. bifida sp.n., and Ch. boettgeri sp.n. Measurements of the genitalia are given in two tables. Material of a mating pair in copulation was available for reconstructing the function of the different genital structures. Mechanical interlocking of these structures can be observed at seven points. Only two of these seven pairs of structures show correlation in shape. Of greater importance for firm interlocking of mates during copulation are the specific distances between the different genital structures. This can be confirmed by simulating copu• lations with models of the genitalia of the three species respectively. This method also allows the identification of conspecific males and females. Since the genus Cheumatopsyche seems to be in a phase of rapid speciation the results of the present study are evaluated under the aspect of the morphological 'Schliissel-SchloB-Prinzip'. Variability of the measurements of the genital structures within specimens which emerged from a small section of a stream makes it doubtful that successful transfer of sperm is possible bctween morphologically extreme mates. In the present case the structure of the genitalia probably is more important as a factor for incipient isolation than pre-and post-mating mechanisms.

Discussion

Ross: If you drew the different values on a chart, would they give you a normal curve? STATZNER: I measured only ]() males and ]() females. ILLlES: We know the paper of REMMERT on Pe!opia (a fly), which occurs in very different sizes but the genitalia are all the same size, just to secure the fitness of the copulation . You find that the variation can also occur in the size of the genitalia without stopping the possibility of mating? STATZNER: In my opinion the successful transfer of sperm is improbable between extreme mates. But complementary investigations are necessary to confirm this hypothesis.

I The whole manuscript will be published in Zoo!. Anz. 193 (5/6).

179 VAILLANT: Can I give another example? It concerns not caddis-flies, but Orthoptera (locusts). In certain large meadows of the Aures mountains in Algeria, many species of the genus Calliptamus live side by side, and, for several of these species, there is a progressive variation of their genital appendages, which have very complicated structures; certain specimens, in that respect, seem to be between two species. Some males do not find, among the many females around them, the suitable one, having the right keyhole for the key made by their genital parts, and cannot reproduce. Considerable variation of this kind may lead to the creation of new specific groups and later to complete genital segregation. STATZNER: Concerning the genus level I think that this does not happen within the Trichoptera. But it may occur on the species level. NIELSEN: The genus Hydropsyche has a similar and rather complicated structure which appears to be associated with glandular tissue; have you studied that? STATZNER: I made preliminary studies of three Hydropsyche species from Northern Germany. Especially in the females I found structures similar to those of Cheumatopsyche. NIELSEN: The structures of Hydropsyche may be homologous with glands found in the polycentropodids. ILLIES: I understand from the paper of TOBIAS that all our German Hydropsyche species can be distinguished in the female sex by the special form of the harpago groove. This must be a worldwide phenomenon? STATZNER: Yes, but the shape of the harpago is, as in Cheumatopsyche, not correlated with that of the harpago receptacle with the exception of H. angustipennis. In Cheumatopsyche a distinct correlation exists between the finger-shaped appendices and the dorsal central part of segment X in the male and the structures of the medial plate in the female. In Ch. bi{tda and Ch. boettgeri the finger-shaped appendices are quite narrow and long and the receptacles show corresponding slits. In Ch. explicanda the apices are rounded and the receptacles are of round shape too. Supplementary information about the medial plates of Hydropsyche females is desirable under the aspect of taxonomy and phylogeny. HANSELL: In regard to differences in genitalia. This is a very efficient means of stopping the hybridization between species. It suggests by implication that any preliminary barriers of hybridization are not very efficient. I wonder if it is known whether there are frequent attempts to mate with members of different species.' FLINT: I know of one case: the genus Frenesia (a North American Limnephilid) which contains two species. Both fly together in the fall, and at one time I placed a large number of both sexes of both species in a bottle. The males became 'excited' for some reason, and became very active, attempting to mate with any other they encountered. There were attempts to mate with males or females of either species. Apparently, the stimulus was enough to release the mating response, but that a lock-and-key device must be needed to prevent incorrect matings. STATZNER: In a letter Mrs. SCOTT wrote that in Cheumatopsyche mates in copula may not be con specific. FLINT: She found the same thing in African Cheumatopsyche? They attempted to mate with other species? STATZNER: She wrote about the phenomenon only. I did not make such observations.

, In May 1975 I captured a pair of Hydropsyche which sat in the usual copulation position on a leaf. The male belonged to H. augustipennis, the female to H. pellucidula. (Statzner).

180 Proc. of the First ent. Symp. on Trichoptera, 1974. Junk, The Hague

A progress report on some approaches to the study of larval house building with particular reference to Lepidostoma hirtum

MICHAEL H. HANSELL

Early in the 3rd instar the larva of Lepidostoma hirtum changes from building a sand grain house of round section to building a house of vegetation panels which is square in section. The panels are almost rectangular and one row covers each of the four sides of the house. Panels cut from dieot leaves such as those of deciduous trees are usually placed on the house with their upper leaf surface facing outwards. Panels cut from monocot leaves, such as grass or reeds are usually oriented with the leaf veins across the long axis of the house. These and other features illustrate the variety of behaviour information contained in the houses of Lepidostoma larvae in particular and caddis larvae in general. Ethology can broadly ask four kinds of question; ones of phylogeny, ontogeny, causation and function. Phylogeny: The rieh behaviour repetoire of caddis larval house building suggests that comparative behavioural studies would yield informa• tion on phylogenetic relationships. Fossil caddis houses offer a rare opportunity to study the fossil record of behaviour yet this field has not attracted attention. Ontogeny: A number of measures of the house building performance of Lepidostoma change over time, some of these changes could be due to internal changes with age; however, bearing in mind the plasticity of some aspects of house building behaviour, the possibility of the learning of skills of caddis larvae should be seriously investi• gated. Causation: This has been the most investigated aspect of house building behaviour. In Lepidostoma it has been shown that hairs at the anterior and posterior of the body regulate many aspects of house building, (HANSELL, 1973, 74). It has been shown that dark/light cycle influences quality of building but no effects of hormone administration on house building have yet been demonstrated. A general feature of caddis house building just starting to be investigated is how a larva whose particle selection apparatus (head and legs) grows in a stepwise manner, manages to construct a house which grows in a curvilinear manner. The preliminary hypothesis is that during any instar the distention of the abdomen is monitored and this informa• tion adjusts the calibration of the selection and building apparatus resulting in a gradual increase in house width and particle size. Function: The mechanical proper• ties of the panel house of Lepidostoma show that firstly it is stronger than an

181 equivalent sand grain house for most types of external loading and that it is very elastic. The elasticity is not due to the properties of the silk but of the vegetation panels; this could be the reason for the preferred orientation of monocot and dicot leaf panels. This house strength and elasticity might serve to protect Lepidostoma larvae from predation by small fish incapable of swallowing whole houses. The house of this and other species may have other functions, such as to protect larvae from wave or current action or conceal them from predators.

Acknowledgements

I should like to thank N. B. BENSON, CAROL NEAL and ANNE ZAJAC for allowing me to refer to experiments of theirs in this report.

References

HANSELL M. H. 1973. Improvement and termination of house building in the caddis larva Lepidostoma hirtum CURTIS. Behaviour 46: 141-153. --- 1974. Regulation of building unit size in the house building of the caddis larva Lepidostoma hirtum. Anim. Behav. 22: 133-143.

Discussion

FLINT: I noticed that you mentioned that the cases protect against predation. There are other suggestions on this utility of the case, such as providing a directed waterflow over the body wall for oxygen transfer. However, the larvae of MageUomyia porteri from Islas Juan Fernandez are interesting in this regard. They construct a case of sand grains that does not cover the entire abdomen, but leaves the anal prolegs free. Perhaps they do not have some important predatory pressure on this island habitat. HANSELL: I was very interested to hear you describing this species which builds a house of sandgrains which does not cover the whole of the larval body length. It sounds as though this would not be effective protection against predation but might improve respiration. ZINTL: Potamophylax latipennis larvae during the penultimate stage with an untouched case do build more by night. HANSELL: That's quite interesting. There was a paper by GALLEPP this year (Freshwater Biology, 1974,4: 193-204) on Brachycentrus which builds a silk tube and feeds by night (it is a kind of filter feeder) and it builds by day. In this instance things are very easy for it because the building material is only silk. This means that it can build a house when it is not doing anything else. In the majority of case building species however the larva must go in search of building material as well as food. So how particular species divide their time between these activities may depend on such things as whether food and building materials are in the same place. It would be interesting to know for a lot of species how they divide up their day. NEBOISS: What about the predation by fish concerning the leaf houses? HANSELL: This must only be speculation but on examining the stomachs of trout we have on occasion found them full of Lepidostoma leaf houses. What we don't know is whether smaller fish are trying to swallow the houses and are unable; they may try to break up the house but eventually give up. An experiment which would be interesting to do in this connection would be to crush houses in an experimental situation such as the one described but with the larva in the house. First this would tell you if the larva within the house adds strength to the structure and

182 secondly it would show the survival of larvae after crushing. But it is an elastic house and it may be that the panel type of construction is used not for structural reasons but for constructional reasons; that is it is more economical to build that way. HIGLER: The cross section of the square house reminds me on the cross-section of plants in which are corners as reinforcements, for example Urtica has this, and it is always for the length elasticity, is it possible that in this case it has the same function? HANSELL: It does not resist very strongly; in fact if loaded in the middle it is less than 20 g, I think. MARLIER: There are several species of cad

183 American genus Pycnopsyche is an ecological equivalent of the European Stenophylax. Studies by Dr. K. W. CUMMINS and by Dr. ROSEMARY MACKAY demonstrate that changes in case• making materials by certain Pycnopsyche species probably are related to interactions between congeneric species living in detrital material. A species may select certain kinds of leaves and bark for cases because these materials are less subject to breakdown by microorganisms, and less palatable to other caddis-flies living in the same situation. Since some of the Lepidostoma species, at least in North America, are also members of the detrital community, there seems to be a possibility that certain of the changes made in selection of case-making materials may also have arisen from the stimulus of other species. HANSELL: Yes. That is quite a possibility. There is some evidence for this in the change from sand grain building to leaf building in Lepidostoma, at least where I have studied it. Although I believe that change-over to be internally and not externally mediated. It does coincide to leaf fall in autumn. But that may be purely accidental. It's quite interesting that Mr. ZINTL'S Potamophylax has the reverse change, it changes from leaf-building to sandgrains, and so one has to find a different explanation for that kind of change. STATZNER: The shelter against predation is one function of the larval case. Another one is its importance for respiration. It is possible, that the change from the sandgrain to the leaf case calises a change of waterflow in the case (managed by Larva's abdominal movements)? HANSELL: That may be true, but the question is then why not build in the style of the leaf house in the first place? STATZNER: The problem of respiration is a smaller one for earlier instars (relation of the surface to the volume of the body). HANSELL: As you can see in the section of the square leaf house, the larva is not using corners. The corners are separated off by a tubular lining of silk. So it is still effectively in a tubular respiratory space. BADCOCK: Does that offer less resistance to water flowing through, as well as efficient respiratory exchange? Or might it be a thigmotactic reaction because the larva 'likes' to have a contact all around its body? It would be touching more of the case in a circular interior than in a square interior. HANSELL: I mentioned this problem but about why it is that the house width goes on increasing when the animal's selection apparatus remains in the same size. This suggests to me that the larva monitors its pressure against the sides of the house. (Slide). This is taken from the DOROTHY MERRILL'S (1965, 1. Exp. Zool. 158: 123-130) paper and I think is quite interesting. The graph shows the head width against larval weight, what you see is that the head width goes up in stepwise manner, but the weights are distributed evenly along the axis. So it means that during anyone instar the selection apparatus remains the same size, but the animal's abdomen is pressing against the side of the house more and more, unless it makes the house wider. This pressure information may be the sensory input which regulates the shift in the calibration of the selection apparatus resulting in the construction of a house of gradually increasing width. DENIS: When a 5th instar larva is taken out of its case, what does it use like materials to make the reconstruction of the first parts of the case? HANSELL: This is given in detail in Mr. SMART'S paper. But briefly: When you take a 5th ins tar larva out of its leaf panel house, and give it sand grains and leaves, it first of all builds a sand grain house. ZINTL: Would you say more about the function of the hairs at the rear end of the abdomen? HANSELL: This will be dealt with in more detail in the next paper. VIGANO: How are joined the two sides of the case with floor or roof at the vertex? Are they joined only with transversal silk? HANSELL: They are joined with silk. Initially a panel is attached to the house by silk strands spun on the inside of the house across all the seams. Afterwards the larva spins a tubular silk lining which cuts across the corners of the house seen in section. However, the silk lining appears to have no structural importance.

184 Proc. of the First Int. Symp. on Trichoptera, 1974. Junk, The Hague

A progress report on the building motivation in the caddis larva, Lepidostoma hirtum

KEENAN SMART

If removed from a 5th instar leaf house a Lepidostoma larva will first reconstruct a sandgrain house before switching to leaf panel addition. The changeover in material selection from sand to leaf is not simply a short-term behavioural switch but a major motivational change involving the appearance of new behavioural items in the animal's repertoire, e.g. cutting and shaping, and the modification of earlier compo• nents of behaviour e.g. fitting and manipulation. The nature of the factors controlling this motivational change were studied. In the example of intact larvae, when the sand-grain section is constructed, certain stimulus conditions are satisfied: (1) Any mechanism measuring building activity will have registered a specific amount of building as completed. (2) The posterior sensory receptors will be covered. It was proposed that a combination of feedback resulting from the satisfaction of 2 criteria: (a) the measurement of a pre-set amount of work completed, (b) positive information from the abdominal receptors on segments 9 and 10 might be regarded as the optimal stimulus which switches the command signal from directing sand-grain selection to leaf panel cutting and introduces panel addition at a characteristic rate. If one (or both) of these conditions is not satisfied, then the switchover stimulus is sub-optimal and the rate of panel addition should deviate from the normal. In an attempt to understand the nature of the internal measure of work done prior to switch over it was further proposed that this might function by monitoring the volume of silk spun during bUilding. Three experiments, based on certain assumptions about the production of silk in caddis silk glands, were designed to test this hypothesis. These assumptions were: (1) In the course of sand-grain construction, spinning of silk depletes the reservoir. If further building is prevented the level of silk in reservoir rises. (2) 5th jnstar larvae living in wild built leaf houses to which only 1 or 2 panels have been added in a month possess a level of silk in reservoir close to or at maximum. In the first experiment larvae were removed from their leaf houses and allowed to construct houses from sand-grains so depleting the silk reservoir to optimal. By delaying the presentation of leaf for 24 hrs, 48 hrs and 96 hrs, it was predicted that the longer the delay the less optimal the stimulus introducing panel addition (since

185 the level of silk in the reservoir will have risen). The prediction that deviation in the rate curves of experimentals compared with controls should be observed was upheld, with rate curves of delay larvae depressed in comparison with controls. In the second experiment, larvae were removed from their leaf houses and placed directly into sand-grain houses. It was predicted on the basis of evidence from experiment (1) and assumption (2) that their rate curve would be depressed. This prediction was upheld. The third experiment examined the effect of interfering with both criteria influenc• ing the normal rate of panel placement. Animals with their anal hooks removed construct sand-grain houses which are longer than those built by intact larvae. On completion of sand-grain building the level of silk in the glands of operated larvae will be below optimal. Sensory information from the anal hooks will be absent. It was predicted that the rate curve would alter from that of controls. Panel fitting rate accelerated in operated animals. (Read by M. M. HANSELL.)

Discussion

DENIS: About the control of the length of the case, I have seen that old larvae of Limnephilus lunatus (few days before pupation) have sometimes a very long case of sand grains (more than two times the normal length). Then the control of the length is not always good. These observations about intact larvae seem like those of MERRILL (1965) about larvae with removed setae and hooks of behind part of the abdomen. But I must say also that larvae of Sericostoma persona tum with setae and hooks removed have not a perfect specific case. This is very important, for this shows that specific building behaviour is controlled by abdominal sense organs in contact with the case. HANSELL: That's right. The only measure of behaviour presented here is one of greater panel attachment. I published some work showing if you remove the 8 long hairs on the last segment associated with the hooks and the 4 hairs on the previous segment, the medial hairs, then it depresses all measures of panel construction: front convex, back concave, side convex; in fact all except for width-greater-than-length which remains constant (HANSELL (1973), Behaviour 40: 141-153). Mr. SMART has followed up this experiment with more sophisticated ones in which he has removed hooks alone, the 8 hairs on the final segment alone, the 4 medial hairs on the previous segment alone, and all the hairs, but leaving the hooks. The control of house length is almost exclusively determined by the presence or absence of the hooks alone. The length of leaf houses with hooks alone removed is very little different from the length of houses with hooks and associated hairs removed. NIELSEN: You mentioned that it was the removal of the anal claws which was the decisive factor. Each claw carries 8 sensory setae. These setae probably are responsible for the reaction. HANSELL: In Dr. FLINT'S talk we heard of this caddis larva that builds a house to cover only the front half of the body. It is hard to see how in this jnstance the anal hooks could have any role in controlling house length which illustrates that control of house length may be brought about in different ways. It would be interesting to have a look at this particular species.

186 Proc. of the First Int. Symp. on Trichoptera. 1974. Junk. The Hague

House building: problems about the spontaneous change of the architectural style in the larva of Potamophylax latipennis ( CURT.) (Trichoptera, Limnephilidae) 1

HERIBERT ZINTL

Introduction

Many larvae of the Limnephilidae and some of the Lepidostomatidae and Brachycentridae were found to use another building material during the 3rd, 4th or 5th instar by which a change of the case style takes place. This change is not induced by the environment, the larva changes spontaneously. Up to now the circumstances of such a spontaneous change have been examined more in detail only for Ounoecia irrorata (SATTLER, 1957) and Lepidostoma hirtum (HANSELL, 1972, 1973). Here a preliminary paper about inquiries on the style change of Potamophylax latipennis announced in ZINTL (1970) is presented. In Fig. 9 (p. 198) two panel cases (Ist style), two transient cases and one almost completed case built of small particles (2nd style) are to be seen. The panel case consists of a dorsal and a ventral row of discs cut out of dead leaves. Its low flanks are built of some particles of leaves. The style change begins by fastening - for our experiments we define: at least two - small particles to the anterior dorsal disc. Some larvae start the change with small particles of leaves or with both sand grains and particles of leaves. But if sand is available all larvae finally build a tubular slightly curved house of sand grains. The style change is conn~cted with the loss of the capability of working with panels. A larva mostly completes only one style change either during the 4th or the 5th instar. In this short paper the following problems will be treated: (1) the distribution of the building activity during the 4th and the 5th instar dependent on the length of the instar, (2) the connection between the lengths of the 4th instar and the time between the 3rd/4th moult and the style change (SC), (3) the connection between the length of the 4th instar and the type of the change (start with sand grains or with small particles of leaves), (4) ideas of the shifting of the motivations for building based on experiments in which the larvae were only denuded or denuded and put into specially constituted tubes of plastics.

1 Devoted to the memory of Prof. Werner Jacobs, Munich 187 Methods

The larvae were collected as 3rd instar at the earliest in October but in most years not until in January. Then each larva lived in a small container of plastics filled with about 300 em' water. The bottom of these containers (80 cm2) was not completely covered with sand grains (diameter 0.6±0.3 mm) and pieces of maple (Acer pseudo• platanus) leaves. Only a small sample of larvae did not get sand. The containers were placed on the steps of a stand like stairs in order to supply them with oxygen by dripping water from one stair to the next below. The constant conditions were: LD 12: 12; white light about 400 Ix 7 a.m. to 7 p.m.; 14± 1°C, 1973 May on 13: 12; light 6 a.m. to 7 p.m. The moults and the alterations of the larval houses were recorded at 6 p.m., 1973 additionally at 7 a.m. As arbitrary units of the building activity we define: (1) panel building activity = addition of 1 disc, (2) small particles building activity = addition of a 1 mm wide ring of small particles (of leaves or sand grains).

Results

1. Building activity during the 4th and the 5th instar There are several different ways of presenting the building activity during the 4th and the 5th instar. As both the lengths of the same instar (Fig. 3, p. 191) and the distributions of the building activities from larva usually differ widely, like other authors I adjusted the building activity sequence of the single larva to the mean lengths of the instars and presented it as the sum of all insects of this experiment. The different distributions of the lengths of the 4th instar are mainly dependent on the time when the larvae were brought under laboratory conditions. In Fig. 1 we have to take notice of the different numbers of larvae of the single years available for these graphs. Only in 1973 were there data up to the pupation enough to plot the small particles building activity during the 5th instar too. As to the panel building activity we perceive that if the ins tar is short it is more probable that bouts of building activity are already in the beginning of the instar. Between the small particles building activity of larvae in a 'sand leaf environment' and a 'leaf only environment' is a big and specific difference. In the incomplete environment the building activity is by way of comparison small, not until the last third of the 5th instar does it begin to increase and even to surpass the data of the larvae living in the complete environment. We can examine the relation between the length of an instar and the distribution of the building activity also by another way. The periods after the 3rd/4th moult of all bouts of building activity are calculated. Then we inquire about the relation between the mean period of all bouts of all larvae with the same length of the 4th ins tar on the one and the length of the 4th instar on the other. In Fig. 2 we see the regression lines based on the data gleaned by this procedure. The mean period of

188 30 Frequency of 20 panel building activity.

particles b. a. 0

10 IV 36 · ". :J L~Jht,.~~ ~w 1 , IV 39 V ~~ .r 102 __ ~~-~:~__ )~' _7_3_~ _~ -..:-'~ : U":....r)l , IV 39 V 102 '70w ,..111, _.,. • ,_, - ~ _""L.• _ IV 7'3 (days) Mean length of the 4th resp. 5th instar

Fig. 1. Frequency pattern of building activity adjusted to the mean length of the 4th and 5th instar; Particles b.a. = Small particles activity; IV = 4th instar; V= 5th instar. Absolute mag• nitudes: 1968: 81 larvae; 1973: 18 larvae; «('73)) = leaf only environment: 14 larvae; 1970:7 larvae.

the bouts plotted against the length of the 4th ins tar shows only a slight increase of the periods for median lengths of the instaL The thickly drawn curve is only a hypothetical one combining the regression lines. We state that dependent of the length of the 4th ins tar there is a shifting of the pattern of the panel building activity during this instaL

2. Period between the 3rd/4th moult and the style change dependent on the length of the 4th instar

Can we also expect a correlation between the length of the 4th ins tar and this special change of the building activity called 'style change'? Would this even be true of larvae which does not change its style in the 4th but in the 5th instar? In Fig. 3 (p. 191) we see both the data of the length of the 4th instar and the data of the style change (days after 3rd/4th moult) to be widely scattered and not normally distributed. Distinct maxima or periodic maxima are missing. If we examine the correlation between the length of the 4th instar and the period between the SC and the 3rd/4th moult for both SCs before and after the 4th/5th moult (Fig. 4, p. 192) it is strange that we only get a significant positive correlation (P = 0.01) for 1973, «1973» and 1967, that is only for mean lengths of the 4th instar of about 40 (median lengths 34 to 38) days. The data of the instar lengths of the

189 Mean panel b. a. during the 4th instar (days after moult 3rd/4th)

'68> 0 '73 20 :' r51@~ - .G,<:~~~. " ~ 0 .". '68" .~~~: O' .. ; ~: ,~GJ: :. . . . ./ . .: 20 40 60 80

'70 /

40

'68'/ ,, 20 ,

Moult 4th/5th

20 40 60 80 (days after moult 3rd/4th) Fig. 2. Correlation between the length of the 4th instar and the mean distance of all bouts of panel building activity to the moult 3rd/4th The bouts of all larvae with the same length of the 4th instar were pooled. Regression lines for several samples, for 1968 and 1973 also the clusters of points. Data of 1968 interpreted with two regression lines. Further explanations in the text.

190 sc 0 1 lL..--,F;-;rfl1TTe"l1Tqurrrne,W,I,n-L,c....,l~l..Ilj, '1I.IIIl.''''.ur'1'..ilIIlwlI ' fUI"'l...... J.Jjdll....' ..1....----1-_--,-.6_8__ _ - "11' 11 "111 • '73 II ," 51 "" I",Lu,!!, " " J. ,

l~~~~II , ~ ~I'~II,, ~I'jI ~I~!" ~,~:I~I ____((~'7_3_») ___ '67 1 I 1,1 'I 'I I , " I 'I 1'''1' '70

1~in" ~ilnii~Ij'i~ITI ~I TI TI~iilrl~' i ,'41~i!TI--~iJ~' ht~I ~II ~I ~' ~' ~" ,~!I ~'~ii~"~1 ,,~t1~----- '69

1I ! jl I I! Ii 40 80 ~~~~thl~--~dl~I"lt~II.lI ~"I I~I,, ~I ,,~,, ~I ~~I~, ----~·6-8---3-1-- 1 '73-39

1 (('73- 39)) 1 '67-42 '70-69 1 11 I 1"111,,, ,I 111 1111.1111 11 1 ! ! ! ' i , ! ! i i 51~ ____ ~I I , ~" ' I~.!,~,I" ~rI,,,~lllI a" "aII I IL1 1 ~l l j l~IIJ~I I II~I W!!II ~III ~I.I~,,' ~:~u1 ~~1 -~~7~1 ll-!I 40 80 120 (days after moult 3rd/4th) Fig. 3. Frequency patterns of style change (SC) and 4th/5th moult related to 3rd/4th moult; Frequency: Absolute magnitudes. Beside the reference number (year) of the experiment the mean length (1968 median length) of the 4th instar. Median lengths from '73 to '69: 36, 34, 38, 68, 74 days. SC data below the horizontal axis mean SC before the moult 4th/5th instar.

191 larvae of these years are in contrary to those of 1970 and 1969 scattered only a little (Fig. 3, p. 191), but this is also true of the insects of 1968 which shows no significant correlation. But if we look for correlations separately for SCs before and after the 4th/5th moult, we can prove correlations for all years except for 1968 and here only for SCs before the 4th/5th moult. The curve of the ratio 'number of SCs post moult to number of SCs ante moult' shows that only in 1973 (but in this year for both the types of environment) nearly all larvae made their SC after the 4th/5th moult. Is it possible that connected with

N

ICoefficient of correlation

Moult 4th/5thIISCp. I IT " IISCa o / / ',0 '0 " IISC c /; 1\

't t't t" i t 20 '68 '73 '67 60 '70 80 (days) '69 Mean length of the 4th instar

Fig. 4. Relations between the mean length of the 4th ins tar and the correlations '4th/5th moult // style change post and ante respectively post or ante moult 4th/5th'. Relations between the mean length of the 4th instar and the ratios 'Number of style changes post / number of style changes ante 4th/5th moult' and 'Mean distance to 4th/5th moult of post moult style changes/mean distance to 4th/5th moult of ante moult style changes'. Absc.: Mean respec• tively median length of the 4th instar above the reference numbers (years) of the experi• ments. Ord.: Coefficient of correlation respectively ratios. SCa = Style changes ante 4th/5th moult; SCp = SC post 4th/5th moult. SC = Sum of both. Marks included in brackets = Dates of 1973, only leaf environment. Cancelled marks = No correlation.

192 specific lengths of the 4th instar a suppression of the SC during the 4th instar takes place? I cannot explain these facts up to now. The curve of the ratio 'mean period' between the SCs post 4th/5th moult and the 4th/5th moult to mean period between the SCs ante 4th/5th moult and 4th/5th moult indicates preliminarily only that is not maintained by the 'post' and 'ante style changes' in relation to the 4th/5th moult a specific distribution. Let us advance in plotting for each larva of a year's sample the SC data - separated in SCs ante and post 4th/5th moult - against the length of the 4th instar (naturally only for such cases in which we got a correlation). Then the regression lines are worked out. Fig. 5 shows the results for 1968 and 1973. In 1968 the scattering is SC (days after moult 3rd/4th)

'6Sa o , p.; '730 0 , . .. p. I • so (('73~ , » I . • (( ,,~

,; • • II 40 .•• . ,; ... • .... . 0

o o o Moult 4th/5th

40 SO (days after moult 3rd/4th) Fig. 5. Correlation between the length of the 4th instar and the point of time of the style change. Regression lines and clusters of points. a = SC ante 4th/5th moult; p = SC post 4th/5th moult. Included in brackets: Sample in only leaf environment.

193 sc 160

120 '69110

80

40

Moult 4th/5th

40 80 120 160 (days after moult 3rd/4th)

Fig. 6. Correlation between the length of the 4th instar and the point of time of the style change including the spontaneous 2nd changes 1969II = 2nd changes of 1969. a and p like Fig. S. Further explanations in the text.

large, in 1973 relatively small. The regression line for '1973, leaf only environment' is significantly (P = 0.001) steeper. In Fig. 6 the regression lines of the results of all years together are presented. We see two separated bundles of regression lines, the one for changes ante moult and the one for changes post moult. 196911a and p mean 'second spontaneous SC of the same larva'. 1969 these processes were rather frequent: after the first SC some larvae relapsed to the panel building activity and completed a second SC. There is a significant (P = 0.01) difference between the regression lines 1969a and 196911a on the one and 1969p and 196911p on the other. Therefore we must conclude that before the 4th/5th moult the changes II are later than the first changes and after the 4th/5th moult the changes II are earlier than the first changes. The thick drawn curves are again hypothetical ones combining the regression lines without 196911. These curves are partially almost straight. These straight parts are based on functions for straight lines reckoned by the following approximation: The regression lines, especially those ones of 1973 and

194 1970 are focused. Their functions are now expressed with the ratio 'period between 3rd/4th moult and SC to length of the 4th instar' always assigning the magnitude 1 to the period between 3rd/4th moult and Sc. Following ratios emerged: for SC's ante 4th/5th moult 1.3:1 up to 1.5:1, for SCs post 4th/5th moult 0.56:1 up to 0.67:1. The ratios based on the regression lines with the smallest scattering of its clusters of points are 1.3: 1 respectively 0.67: 1. It would be possible to interpret these facts with the terms of the oscillation theory, but it must be found out what oscillation is here. This might be especially difficult relating to the motivations for the building activity. But let us hypothetically interpret in spite of these problems: If the ratio 'duration of the period of the triggering oscillation(s) (probably cycle of moults) to the duration of the period of the triggered oscillation(s) (probably motivation for SC)' is about 1.33: 1 (SC ante 4th/5th moult) or 0.67: 1 (SC post 4th/5th moult) entrainment (=Mitnahme) results (basic principles: ASCHOFF & WEVER, 1962).

3. Type of style change dependent on the length of the 4th instar and the period between the 3rd/4th moult and the style change It is interesting that there is also a relation between the type of the SC process and the length of the 4th instar. From the graphs of Fig. 7 (p. 196) we can gather that connected with short to middle lengths of the 4th ins tar during the time of the ascending frequency and during the maximum of SCs change processes with sand grains predominate. We might imagine that the shifting of the motivations in the central nervous system in insects with a shorter instar length is more rapid and leads at once to the building activity with sand grains.

4. Shifting of the building motivations during the time round the style change About the shifting of the motivations we got pointers from the following experi• ments: Larvae living in a panel house (4th and 5th instar) were denuded every three days; other larvae were denuded in the same sequence but starting 20 days after the SC (SC in the beginning of the 5th instar). In another samples the larvae were denuded only once 1,4, 7, 12 or 18 days after SC. After the denudation the larvae display three possible reactions: (a) No building, (b) Building with panels in a transitory way after having established the pre-case, (c) Adding small particles to the pre-case at once. The results of these experiments are to be seen in Fig. 8 (p. 197). In order to describe the behaviour after denudation we use two ratios: Relative building activity = number of building larvae/number of all larvae of this experiment. Relative panel building activity = number of larvae building with panels/number of all building larvae. The graph (a) and (b) show that from these insects denuded between the end of the 4th and the beginning of the 5th instar not only did an increasing number not build a house but also an increasing number from those having built a house did not use panels. Graph (c) demonstrates the possibility of triggering off a relapse to the panel

195 SC classified in types and pooled in blocks of 20 days

10

'67

.....

20

10 '68

J------O ..;

20

10 '73

t------Q-----i Moult 4th/5th 60 120 (days after moult 3rd/4th)

Fig. 7. Relations between the type of starting the style change, the position of this change after the 3rd/4th moult and the length of the 4th instar. Black blocks: Start with leaf particles. White blocks: Start with sand grains. Striped blocks: Start with mixed particles. Half filled square marks: Sum of the SC within 20 days.

196 ·...... -

r ·_·v-,r..... \.~A~,-- \i\ ~, \ . 'I--\ \ \ \ '-. \ \ 60,4 \ \ .0 \-_. \ \ \ \ \ .~O,2 .\ .. "0 \ Qj a) b) c) \~,I d) 0::- , • ,. • ,...... ,--.--,...-.-, -'IlL' ..... T"""r-"--'---' 1. 2. 3. 4. 5.6. 7. 8. 9. 1. 2. 3. 4. 5.6. 7. 8.9. 1. 2. 3.4. 5.6. 7. 147 1218 denudation days during the 4th i. extending after SC beyond 4th/5th m. denuded (Start before SC) Start 20 d. after SC Fig. 8. Relations between the tendency to build and the tendency to build with panels around the time of changing the architectural style Absc.: Number or day of denudation. Ord.: Explanation in the text. style even 32 days after the Sc. The tendency to build a house at all increases after a transitory decrease. Against the conclusion that the mentioned curves might only indicate the exhaus• tion of the tendency to build by denudations we refer to the decreasing trend of the curves of graph (d) although each larva was denuded only once. We also state here a decrease of the tendency to build with panels. We cannot expect the increase of graph (c) in graph (d) because it might appear only about 32 days after Sc. Based on these results we draw the conclusion that the threshold for building with panels rises very slowly, for many days. If this process is advanced only a consider• able increase of the common motivation for building might trigger off working with panels. The strong stimuli of denudation increase this common motivation for building and so induce a transgression of the already rather high threshold. It is possible to make some statements of the meaning of special stimuli for the transgression of the panel building threshold. For this purpose larvae were pushed out of their houses only 1 day after Sc. Then they were put into a substitute case of plastics. Following conditions were combined in different ways: case transparent or opaque, of natural length or half the natural length, light regime DD or LL (600 Ix). We compare the magnitudes of the relative panel building activity deduced after 3

197 Fig. 9. Change of the architectural style of the case 2 panel cases (4th instar; laboratory), 2 transient cases and 1 little particles case almost completed (5th instar; collected in the wild). Maple leaf (Acer pseudoplatanus) with spots of Rytisma acerinum. The dead leaf including this Ascomycete is food for the larvae. Disc cut out by a 5th instar larva: 10 mm diameter.

Substitute case 00 LL Light dep. incr. (b) (a) (a- b) nude 0,58 0,87 0,29 0 0,28 0,56 0,28

~ 0,10 0,26 0,16 • (0,28) 0,76 0,48 (0,10) 0,38 0,28

Fig. 10. -Data of relative panel building activity depending on conditions of stimulation after denudation White marks: Transparent cases. Black marks: Opaque cases. Length of the marks: Half or complete natural length of the case. Further explanations in the text.

198 days (Fig. 10, p. 198). It is understandable that the magnitude of 'nude' under LL' is the highest, the magnitude of 'plastics case of the normal natural length under DD' is the lowest. But it is very strange that the amount of increase dependent on light of the relative panel building activity is rather high if opaque plastics tubes of half the length under LL are used. This fact is probably caused by a specific behaviour of the larva under such conditions: it slides to and fro and so it receives strong new stimuli due to the changes in intensity of light which it experiences and the changes of body contact with the case, for instance between anal hock and case (DIEHM, 1949) (MERRILL, 1965), perhaps also due to the changing contraction and expansion of the abdomen (stretch receptors: OSBORNE & FINLAYSON, 1962). It seems that the increase dependent on light (Fig. 10.) is composed of units of about 0.15 Relative panel activity. But this has not yet been proved because the scatterings are almost completely missing.

An attempt to interpret

It is too early to compose a complete theory of the spontaneous style change, but some hypotheses as a basis for such a theory can be put forward. I want to differentiate between a common tendency to build and special tendencies to build with panels respectively to build with small particles. These ideas are supported by the results shown in Fig. 8 (p. 197) and Fig. 1 «'73)) (p. 189). The common tendency to build is very different from larva to larva. But the bouts of building activity are not distributed only by chance. There are higher probabilities for bouts dependent on the length of the instar (Fig. 1, p. 189; Fig. 2, p. 190) and on the diurnal rhythm (not referred in this paper). The tendency to build with panels decreases over a time up to two weeks and more (Fig. 8, p. 197). Perhaps this decreasing tendency (or the common tendency to build) oscillates. This would explain the spontaneous relapses to the panel building activity and then the second SCs (Fig. 6, p. 194). There is a correlation between the length of the 4th ins tar and the period between the 3rd/4th moult and the SC, excluding very short instars (Fig. 4, p. 192; Fig. 5, p. 193; Fig. 6, p. 194). Two points of time of the SC are possible: the one before the 4th/5th moult, the other after this moult (Fig. 6, p. 194). The reason why some larvae use the early gate, while others use the late gate of time is unknown. 1973 (Fig. 4, p. 192; Fig. 5, p. 193) only a tiny number of larvae changed their architectural style before the 4th/5tp moult. Probably there are many periods of time for the ripening of the tendency to make the SC, but there are only two gates for starting the SC: the one before the 4th/5th moult, the other after this moult (compare: 'gaJed and ungated events of the development of Drosophila .. .'; PIT• TENDRIGH & SKOPIK, 1970). The gates may have as background especial differences of the phase angle between two or more oscillations. Also another correlation was proved between the length of the 4th instar and the SC and that is the one with the type of the change. The shorter the 4th instar the

199 more the data of the change are concentrated and the more frequent are changes at once with sand grains (Fig. 7; '67, '68, '73; p. 196). We suppose the shifting of the motivation then to be more advanced. For this case the hypothetical oscillation of the tendency to build with panels is damped more strongly, compared with the conditions of a very long instar (Fig. 7; '70; p. 196) when the amplitudes of the tendency to build with panels decrease only slowly, visible with many SCs not started with sand grains. If after the SC sand is not available the rate of building decreases. This is a pointer to alterations of the 'innate releasing mechanism' in the direction to the material 'sand'. The most interesting result of the denudation experiments combined with artificial houses is the assumption that the common motivation for building can be increased probably in the form of small regular steps. But this hypothesis is to be further examined.

Summary

1. The arrangement and the amount of the building activity during the 4th and the 5th instar (also in an incomplete environment) were examined. 2. During the 4th or the beginning of the 5th (last) instar all larvae change spontaneously from a leaf to a sand house. 3. Apart from quite short 4th instars there is a positive correlation between the point of time of the style change related to the 3rd/4th instar and the length of the 4th/5th moult but only if changes before and after the 4th/5th moult are interpreted separately. 4. For this correlation a curve and an oscillatory interpretation are proposed. 5. There is also a correlation between the type of the style changes, the distribution of the points of time and the lengths of the 4th instar. 6. Experimental denudations support hypotheses about the shifting of the motiva• tions for building.

References

ASCHOFF, J. & WEVER, R. (1962): Biologische Rhythmen und Regelung. In: Probleme der zentralnervosen Regulation; V. Bad Oeynhausener Gespriiche. Berlin-Gottingen-Heidelberg, Springer-Verlag. HANSELL, M. H. (1972). Case building behaviour of the caddis fly larva Lepidostoma hirtum. J. Zoo!. Lond. 167: 179-192. --. (1973): Improvement and termination of house building in the caddis larva Lepidostoma hirtum CURT. Behav. 46: 141-153. --. (1974): Regulation of building unit size in the house building of the caddis larva Lepidostoma hirtum. Anim. Behav. 22: 133-143. MERRILL, D. (1965): The stimulus for case building activity (Trichoptera). J. Exp. Zoo!. 158: 123-130. OSBORNE, M. P. & FINLAYSON, L. H. (1962): The structure and topography of stretch receptors in representatives of seven orders insects. Quart. J. Micro. Sci. 103: 227-242. 200 PITIENDRIGH, C. S. & SKOPIK, S. D. (1970): Circadian Systems, V. The Driving Oscillation and the Temporal Sequence of Develoment. Proc. Nat. Acad. Sci. 65: 500-507. SATILER, W. (1957): Beobachtungen an den Larven von Crunoecia irrorata CURT. (Trichopt• era). Ber. limnolog. F1uBstn. Freudenthal 7: 18-22. ZINTL, H. (1970): Zum Problem der Wahl des Baumaterials bei der Larvae von Potamophylax latipennis (CURT.) NEB. (Trichoptera, Limnephilidae). Z. Tpsych. 27: 129-135.

Discussion

NIELSEN: I think you have worked with Potamophylax nigricornis rather than latipennis. I have described the house building of P. nigricornis; it will start with a case made of fine sand grains, go over to a flat case of beech leaves and end with a case of coarse sand grains. The case of P. latipennis is quite another piece. It starts with various vegetable matter and goes o"er to sand grains. (After my return to Copenhagen I have went through a rather large collection of larvae of Potamophylax latipennis from forest brooks in Sjaelland. In this I have found some larvae with a flat case built of shed beech leaves like those of P. nigricornis, though not as broad. Hence, I think that ZINTL is correct in referring to his material as P. latipennis. From streams in Jutland I know no latipennis cases of this type). ZINTL: I can't say much. Mr. DOHLER had determined my animals as latipennis. HILEY: In Britain P. latipennis, P. cingulatus and Chaetopteryx villosa all start in the 1st instar with sand grain cases, then they often change to some form of leaf case, sometimes as you described, sometimes without the large pieces of leaves, sometimes with long sticks. In the 4th or 5th instars the three species change back to mineral case building. In Potamophylax the final instar has a very definite case, made of small stones with larger ones along the sides. Chaetopteryx has a more variable case, usually made of mineral, but in no definite style. ZINTL: In Germany P. latipennis during the 2nd ins tar also has the change of building material you described. But I only studied the late style change. This late change is sometimes found twice. I don't know why. In such years I collected the insects in November, and took them into the laboratory. Perhaps this is a reason for this fact. HANSELL: It seems to show that the relationship between the position of the moult and the style change is not produced in a simple way. I think these results go some way towards explaining my own failure to experimentally induce style changes from sand grain to leaf and vice versa in Lepidostoma by hormone administration. These were based on a rather simplistic assumption that style change was controlled by a simple hormonal trigger. The results shown here suggest that style change may be regulated in a more complex way and that it may be possible to get a lead on the nature of the mechanism by looking at the length of the instar in relation to the style change.

201 Proc. of the First Int. Symp. on Trichoptera, 1974. Junk, The Hague

A progress report on the locomotion behaviour of a larva of Limnephilidae (Micropterna testacea) in water currents

MICHEL BOURNAUD

This case larva walks on exposed biotopes of streams up to various stream velocities (often until 100 cm/s) which produces on it a push reaching 30 times its weight. I made a comparison between the effect of a single component (i.e. the force, experimentally separated from the current) and the effect of the current as a whole upon the locomotory behaviour of the animal. In these two conditions, the behaviour is composed of a passive orientation by vane effect, a kinesis (measured by the power developed) and a pathy (measured by the proportion of individuals which resist a given force or current). The most part of the locomotion is made against the current or the force (positive orientation). Negative locomotion practically stops at 50 cmls or 1200 mg and above. A study about kinematics and dynamics of the locomotion will be published in the journal Hydrobiologia. I have studied the locomotion of this larva by means of films. This locomotion follows a sequence similar to that one usually observed in walking insects, with protraction of legs from behind to forward. The walking is slow. The locomotion includes phase A (advance of the body) and phase B (case towing, with protraction of posterior legs). These two phases perfectly agree with the basic pattern described by DENIS (1968) for the larval case construction. In a current, the sequence of protraction is often modified by supplementary protractions of the anterior and middle legs (exploration or skidding). The current induces a decrease in the mean locomotion speed by shortening and chiefly separat• ing the towing phases, which are the only times of rapid advance of the case. So when current speed increases the case advance (phase B) does not change very much in velocity, but gets rarer and rarer. Therefore locomotion speed (phase A + B) becomes lower. The energy expenditure is as follows: (1) Only a small part of the energy brought by the metabolism is used in global locomotion. (2) The whole energy is used in phase B (advance of the case). So the metabolism increases in current. (3) All these values are lower than those found before in animals, and the mean traction strength of M. testacea is very high (up to 60 times of the larval weight). Therefore this 203 species uses more energy in traction than in speed, the opposite to the situation is fishes. It is an ethological adaptation to life in a current, reinforced by the necessity of pulling a case which increases the force of the current.

Discussion

HANSELL: I am interested in the flexibility of the stepping pattern. I wonder whether the pattern of movement of the legs can be divided into a small number of patterns or whether there seems to be a great range of possible stepping patterns. The reason I ask is that, just from looking at walking and the manipulation of particles by caddis-larvae, it is surprising to see the extraordinary lack of any stereotypy in the movements of the legs.

BOURNAUD: I think there is a hazard in the progression because of the irregularity of the force of the current, they slide on the ground, and they need a supplementary movement of the anterior legs to resist and to retain themselves after sliding. Only the posterior legs have a regular pattern of equilibration and move at the same time when the case moves.

MORETTI: Je voudrais savoir si les larves tres jeunes se comportent de la meme fa~on; si vous avez plus d'experience et si vous avez un deuxieme pro jet en train.

BOURNAUD: Je n'ai pas fait d'experiences pour les jeunes larves parce qu'elles sont trop petites et j'ai utilise des methodes photographiques qui donnent des resultats aleatoires sur les jeunes larves.

204 Proc. of the First Int. Symp. on Trichoptera, 1974. Junk, The Hague

A progress report on die) rhythmicity in Trichoptera

JOHN o. SOLEM

Diel locomotoric rhythmic patterns in the larvae and adults of Agrypnia obsoleta HAG., Athripsodes aterrimus STEPH., Molanna angustata CURT., and Limnephilidae varia are reported. The field investigations were carried out in a lake near Trond• heim, Norway, situated 63°N and IO°E. In the lake the larvae were collected by traps at two-hour intervals throughout several 24-hr periods, covering 19 months. Behaviour of the adults like emergence, swarming, oviposition, etc. were also observed continuously during both days and nights. Climatic conditions were meas• ured. The three species mentioned and Limnephilidae varia did not show any clear diel rhythmicity during wintertime, which may be due to small numbers collected or that the arhythmicity is a normal pattern. The diel periodicity was most obvious in spring/summertime - the time just before pupation. In Athripsodes aterrimus and most likely also Molanna angustata the diel locomotoric rhythmicity of the larvae and the adults had the same phase, a day activity. In Agrypnia obsoleta a clear activity phase of the larvae was found at daytime, around midday, while emergence and swarming of adults occurred around midnight. Exactly when this change in activity phase or change in response to the Zeitgeber occurred, is from the present results difficult to state, but it is very likely that it was during pupation. In spring/summer the larvae of Limnephilidae varia had a clear day activity. To my knowledge the various Limnephilidae-species in the collection emerge, swarm, and fly at nighttime. The behaviour of Limnephilidae varia thus seems to be very much the same as that of A. obsoleta. As final remarks it can be pointed out that at least two different locomotoric activity patterns have been demonstrated'.

1. The larvae and the adults have the same phase of locomotoric activity. 2. The larvae have one phase of locomotoric activity and the adults the opposite phase - which means that from the larvae to the adults the response to the Zeitgeber is changed 180°.

205 Discussion

CRICHTON: I am not clear how you work out the adult activity. SOLEM: I knew the area and I knew the place where emerging, swarming etc. took place. In studying the activity I was on the place and observed throughout a number of days and nights at what time during a 24-hr period the different kinds of activity was performed. As I used no traps I had to make my own scale for reproducing the data in a figure. It was easy to state when there was no activity. Then I found it appropriate to divide the activity period into one period including the starting/beginning and the ending, and another period when e.g. mass swarming (activity) was performed. The limits between the different periods were when more or less than 10-15 individuals were seen. DENIS: Do you know where the specific behaviours (feeding, building ...) of the larva take place in the day cycle and in the larval instar? Indeed each adult behaviour has a determined place in the day cycle. For instance each period of activity begins in afternoon for several species of Limnephilidae in laboratory. Mating behaviour and oviposition occur at the begin• ning of afternoon for Anabolia nervosa and during night or early morning for several species of Limnephilus. SOLEM: Concerning feeding and building behaviour of the larvae, I have no idea, as they were captured in traps, but the behaviour of the adults was thoroughly studied. For Agrypnia obsoleta, for instance, I know that emergence and swa!ming, and partly also copulation, occurred at night-time. Copulation did also occur at day-time, and so did also, at least, some of the oviposition. MARLIER: For a number of genera copulation and oviposition in the laboratory is known during the night, and the animals are always in copulation in the early morning, but for Anabolia nervosa the copulation occurs at the beginning of afternoon, and oviposition at the same time. SOLEM: In field observations on Agrypnia obsoleta, I found that activity was dependent on temperature. The adults had one behaviour pattern below 10°C, another between 10-14/15 °C, and a third one when the air temperature exceeded 14/15 0c. When the temperature exceeded 14/15°C, swarming occurred, and together with swarming copulation was performed. In the temperature interval 10-15°C there was no swarming, but copulation occurred during the whole 24-hr period. Below 10°C the animals were hiding in the vegetation and very difficult to observe, and I do not know if they copulated or not. But there was a distinct change in the behaviour at the temperature level of 14/15°C. MORETTI: Do you have any sort of bait in the traps? SOLEM: No, empty traps only. HILEY: Are the figures for the number of larvae captured averages or totals? SOLEM: Totals.

206 Author index

AHMAD 138, 141 F'INLAYSON 199, 200 AISA 114 FiORELLI 114 ALDERSON 134, 141 FiSCHER 92, 114 AMBUHL 51, 57 FLINT 6,9, 17,42,47,48,57,69,76,85, ASCHOFF 200 104, 142, 143, 146, 170, 180, 182, 186 FOURCROY 39 BADCOCK 4911., 145, 146, 154, 157, 158, FROCHOT 22, 105 184 BAILEY 123 GALLEPP 182 BEAUJOT 171, 175 GIANOTTI 92 BERLAND 92 GOWER 125, 129 BETIEN 175 GRENSTED 55, 57 BILLBERG 34, 39 BITSCH 105 HAAGE 171,175 BOESIGER 61,65,68 HALLER 175 BONE 172,175 HANNA 22 BOTOSANEANU 26,29,30, 41ft., 49, 55, 57, HANSELL 6, 23, 30, 142, 158, 180, 181ft., 59ft., 92, 104, 164 186, 187, 200, 201, 204 BOULARD 64, 68 HENNIG 61, 62, 68 BOURNAUD 24, 42, 146, 160, 203ft. HICKIN 22, 114, 124, 129, 134, 139, 141 BOUVET 105ft. HIGLER 30, 92, 108, 130, 158, 183 BRAUER 76 HILDREW 55, 57 BRAY 112 HILEY 18, 19, 21ft., 69, 115, 129, 130, 146, 158, 160, 161, 165, 18~ 201, 206 CHAMPEAU 112 HORWATH 168, 170 CIANFICCONI 85,92, 9311., 11111., 142 HYNES 7,17 CONAT 114 CRICHTON 6, 31, 40, 58, 76, 124, 125, 129, ILLIES 5, 6, 7, 17, 18, 23, 45, 48, 58, 68, 72, 130, 147ft., 206 75,76,177,179,180 CUENOT 29 CUMMINS 184 JEUNIAUX 171, 175 CURTIS 49 JONES 117ft., 131ft., 157

DANECKER 135, 137, 138, 141 KiMMINS 34, 39 DARLINGTON 16,17 KLAPA.LEK 77, 85 DECAMPS 6, 26, 29 KOCH 172, 175 DENNIS 106, 108, 109, 184, 186, 206 KOMNICK 173, 176 DENNING 7, 17 KRAWANY 173ft. DIEHM 199 KUMANSKI 104, 159 DI GIOVANNI 114 DOBZHANSKI 65, 68, 69 LANGFORD 56, 57 DOHLER 49, 57, 201 LEACH 39 LEADER 171,175 EDINGTON 51, 55, 57, 134, 141 LE LANNIC 108 EDMUNDS 3 LEPNEVA 22, 33, 39, 134, 141 ELLIOTT 6, 165 LHONORE 161 207 LINNAEUS 39 SARETCHNAJA 114 LORENZONI 114 SATUA 138, 142 LUBBEN 175 SATTLER 58, 187, 201 SCHACHTER 114 MACKAY 184 SCHMID 8,17,68,77,85,92,104,105,114 MALICKY 30, 42, 70, 71f1., 108, 114, 165, SCOTT 180 171,176,177 SEHNAL 106, 114, 149, 157 MARCUZZI 114 SILTALA 171, 176 MARINKOVIC-GOSPODNETIC 49, 55, 57, SKOPIK 199, 201 7711., 104 SMART 184, 185, 186 MARLIER 40, 43, 114, 183, 206 SMUTH, P. VV. 168, 170 MARSHALL 168, 170 SMITH, S. D. 5, 6 MARTYNOV 76 SOLEM 20511. McLACHLAN 49, 57, 92, 93, 104 SOLOMON 131, 142 MERRILL 184, 186, 199, 200 STADLER 174, 176 MICHALON 107 STAl'ZNER 31, 69, 179, 180, 184 MILNE 34,39 SniFFAN 64, 68 MORETII 29,40,76,8711.,9311.,108,11111., STEPHENS 39 160, 176, 204, 206 SUTCLIFFE 171, 176 MORSE 3311., 68, 70, 145, 146 SVENSSON 57, 106 MORTON 139, 141 SZCZESNY 65,68,69 MOSELY 1, 33, 39, 42, 49, 57,92 MURGOCI 68 TATICCHI 112, 114 TAYLOR 147, 157 NAOUMOFF 171,175 THIENEMANN 175,176 NEBOISS 6, 23, 45, 48, 85, 160, 182 TJEDER 57 NIELSEN 5,9,17, 18, 19, 22, 23, 24, 39, 57, TOBIAS 49, 55, 57 69, 115, 130, 142, 157, 15911., 16311., 180, 183, 186, 201 ULMER 49,57 NovAK 106, 114, 149, 157 UNZICKER 169, 170 NUSKE 173, 175, 176 VAILLANT 6, 18, 2511.,68, 75, 76, 142, 177, OSBORNE 199, 200 180 V ANDEL 64, 68 PEDROTII 114 VIGANO, A. 8711. PHILIPSON 51, 57 VIGANO-TATICCHI, M. I. 8711., 184 PICTET 26, 34, 39 PIRISINU 92, 11111. VV ALKER 34, 39 PITTENDRIGH 199, 201 VVALLACE, I. D. 22, 23, 3311. VVALLACE, J. B. 145, 146 RAnovANOVIC 77, 85 VVATSON 50 RAvlZZA 114 VVEBER 169, 170 REICHEL 114 VVEVER 200 REMMERT 179 VVICHARD 114, 17111. RESH 40, 69, 130, 145, 16711. VVIGGINS 3, 6, 711., 23, 29, 42, U4, 183 RETZIUS 39 VVILKENS 67, 68 REYNOLDSON 129 VVILLIAMS, 147, 157 RICHARDSON 168 VVINKLER 114 Ross 1,6,23,42,58,69,70,104,142,143, 168,176,177,179 ZINTL 23, 92, 115, 182, 184, 18711.

208 Subject index

Acer pseudoplatanus 198 Athripsodes 3311., 16711. Acrophylax 6211. A. albifrons 36, 37, 38, 40, 149 aestivation 71, 10511. A. angustus 167, 168 Agapetus 72, 73 A. aterrimus 35,36,37,38,40,11811.,205 A. caucasicus 73 A. bilineatus 38, 40 A. delicatulus 152 A. braueri 40 A. .juscipes 149, 164 A. cancellatus 168 Agrypnia obsoleta 205, 206 A. cinereus 36, 37, 40, 130 A. varia 112, 113 A. commutatus 40 Allocapnia (Plee.) 69 A. cuneorum 40 Allogamus 23, 6111., 90 A. dissimilis 152 A. auricollis 22, 89, 91 A. maticus 168, 169 A. dacicus 6311. A. erullus 168, 169 A. lazarei 6211. A. fulvoguttatus 40 A. mendax 66, 89, 90, 91 A. genei 40 A. stadleri 66 A. inaequalis 40 A. starmachi 6311. A. interjectus 40 A. tatricus 6211. A. leucophaeus 40 A. uncatus 6311., 89 A. longispinosus 40 Anabolia 109 A. menteius 168 A. brevipennis 24 A. nigronervosus 149 A. nervosa 109, 11811., 159, 176, 206 A. resurgens 168 anal papillae 17311. A. rieli 40 Anisoptera 177 A. saccus 168, 169 Annitella 59ff., 73 A. submacula 169 A. dziedzielewiczi 5911. A. tarsipunctatus 168 A. chomiacensis 5911. A. tavaresi 40 A. kosciuszkii 5911. A. transversus 168 A. lateroproducta 5911., Athripsodina 40 A. thuringica 5911. Athripsodini new tribe 34 A. transsylvanica 5911. Austrian Alps 2511. A. triloba 62 Antillopsyche 47 Apatania 18, 23, 72, 73, 90 Barypenthus 48 A. cypria 75 Beraeamyia 73 A. jimbriata 90, 91 B. aphyrte 74 A. intermedia 164 B. hrabei 74 A. muliebris 22, 163, 164, 165 B. kutsaftikii 74 A. nielseni 164 B. schmidi 74 A. wallengreni 22 B. squamosa 74 Apataniinae 18 Beraeidae 29, 72 Apennines 9311., 11111. Brachycentrus 42, 182 Ashmira 43 B. subnubilus 149 Asterophora heeri (Gregar.) 112 Brachycera(Diptera) 177 Astyanax (Pisces) 64 braehypterism 63

209 building activity 135ft., 181ft., 185ft., 187ft. Cymus f1avidus 118ft. building motivation 185ft. C. trimaculatus 118ft. building style change 187ft. deterioration, environmental 167ft. Calamoceras 73, 76 diapause 112, 149ft. C. illiesi 74, 76 Dicosmoecinae 19 C. marsupus 74, 76 die I rhythmicity 205ft. Calamoceratidae 72 differentiation 77ft. Caldra 73 Diplectrona 41, 42 Calliptamus (Orthopt.) 180 D. felix 51, 52, 55, 56 Carex buxbaumi 112 qistribution, geographical 1ft., 25ft., 49fI., Caricetum 111 59ff., 71ff., 77ft., 87ft., 93ft., 147ff. Catagapetus 41 Dolophilus 30 C. nigrans 41 Drosophila (Diptera) 199 cavernicolous Trichoptera 71, 105ft. D. paulistorum 68, 69 Ceraclea 33ft. D. willistoni 68 C. albimacula 37, 40 Drusinae 90 C. alb'oguttata 40 Drusus 18, 23, 73, 77, 89, 90, 91, 93ft. C. annulicomis 36, 37, 40 Drusus annulatus 129 C. aurea 40 D. bosnicus 77ff. C. cancellata 38 D. discolor 89, 90, 91 C. dissimilis 37, 38, 40 D. improvisus 93ft. C. excisa 40 D. klapaleki 77ft. C. fulva 36ft. D. medianus sp.n. 79ff. C. minima 36 D. mixtus 93 C. nigronervosa 35, 40 D. monticola 77 C. nepha 36 D. nigrescens 77 C. norfolki 40 D. plicatus 77ff. C. nygmatica 40 D. radovanovici 77ft. C. perplexa 40 D.r. septentrionis ssp.n. 78ft. C. riparia 40 D. ramae 77ff. C. senilis 37,40 D. verspertinus sp.n. 80ft. C. sobradieli 40 D. taxon 193ft. C. tarsipunctata 36 D. taxon 293ft. C. transversa 36 Dryops (Coleopt.) 64,68 Chaetopteryx 23 Dytiscus (Coleopt.) 183 C. villosa 22, 201 Cheumatopsyche 69, 179, 180 C. bifida 179, 180 Ecclisopteryx 23 C. boettgeri 179, 180 E. guttulata 159 C. explicanda 179, 180 Ecnomus tenellus 118ff., 160 C. lepida 55, 56, 57 Elminthidae (Coleopt.) 176 Chionophylax 62ft. Elodidae (Coleopt.) 176 C. czamohoricus 67 Enoicyla 76, 176 C. monteryla 66, 67 E. costae 76 Chironomidae 43 environmental deterioration 167ft. chloride epithelia 173ft. Eothremma 43 Copepoda 43 Ephemeroptera 177 Crunoecia 23, 183 Eristalinus (Dipt.) 176 C. irrorata 187, 201 Erpobdella octoculata (Hirud.) 142 Cryptothrix nebulicola 89, 90, 91 evolution 30, 59ff. Culicidae 176 extinction 159ft., 167ff. 210 Farula 42 H. angustipennis 51, 54, 56, 154, 155, 157, feeding nets 145, 146 158, 159, 160, 180 fine structure 173ft H. contubernalis 55, 56, 57, 158 fish-farms 159fl. H. discreta 75 Fontinalis 183 H. dissimulata 160 food of larvae 7t1., 42, 138, 145, 167t1. H. exocellata 56 France 25t1. H. fulvipes 55, 56, 57 Frenesia 180 H. guttata 56, 57 freshwater sponge 130 H. instabilis 49, 55, 56, 57, 73 functional morphology 179 H. omatula 57 H. pellucidula 51, 56, 180 genitalia lOti., 25t1., 33t1., 59t1., 77t1? 93t1., H. saxonica 55, 57 179 H. siltalai 49, 51, 53, 55, 56 glacial relicts 163 Hydropsychidae 49,72 Glossosoma boltoni 176 Hydroptila angulata 160 Glossosomatidae 176 H. kumas 72 Glyphotaelius 109 Hydroptilidae 29,72, 175 Goera 16, 18 Goeracea lOti. Imania 8 G. genota 13, 16 Isopoda 61, 64 Goereilla baumanni 7t1. Goeridae 175 Jutland 159, 163 Goerita lOti. Goerodes 183 Kokiriidae 47 Grammotaulius atomarius 112, 113 Great Britain 21t1., 49t1. Lake Erie 167t1. Gregarina limnophili (Gregar.) 112 larval house building 41, 42, 135t1., 145, Gregarinida 112 181t1., 185t1., 187t1. Grumicha 48 Lasiocephala 23 L. basalis 174,176 Halesinae 90 Lepania cascada 7t1. Halesus 23, 89, 90 Lepidostoma 23 H. digitatus 22, 159 L. hirtum 6, 152, 154, 156, 157, 181t1., H. radiatus 22, 109, 151, 153 185t1., 187, 200 H. rubricollis 90, 91 Lepidostomatidae 18,23,72 Helicopsyche 73, 76 Leptoceridae 33t1., 72 H. bacescui 76 Leptocerus 33t1. H. borealis 1 L. tineiformis 112, 113 H. heacota 1 light traps 147t1. H. iltona 1 Limnephilidae 7, 21t1., 72, 90, 175, 205 H. lusitanica 76 Limnephilinae 90 H. megalochari 76 Limnephilini 176 H. revelieri 76 Limnephilus 72,90, 109, 206 H. sperata 76 L. aflinis 22, 149, 152, 158, 176 Helicopsychidae 1t1., 72 L. auricula 109 Himmerland 163, 164 L. bipunctatus 109, 112, 113 histochemistry 161, 173t1. L. coenosus 89, 90 house building 181t1. L. decipiens 24 Hydatophylax infumatus 22 L. extricatus 23, 90, 91 Hydraena (Coleopt.) 73 L. flaviconis 112, 113, 115 Hydrocorisae (Heteropt.) 177 L. fuscicomis 23 Hydropsyche 49t1., 73, 75, 89, 175, 180 L. fuscinervis 21 211 L. griseus 23 O. albicome 89, 90, 91, 163 L. hirsutus 91 Oecetis lacustris 11811. L. incisus 22 Oligoplectrum maculatum 159 L. lunatus 22, 24, 109, 129, 186 orobiontic fauna 8711. L. marmoratus 11811. osmoregulation 17111. L. minos 72, 76 L. rhombicus 89, 90, 91, 109 Paduniella 73 L. spars us 112, 113, 115 Parachiona 18 L. stigma 89, 90, 91, 176 P. picicomis 163 L. vittatus 89, 109, 112, 113,115 Pararhyacophila 89 Lithax 18 Pelopia (Dipt.) 179 locomotion behaviour 20311. phenology 2611., 10511., 11711., 14911. Lype 130 Philanisus plebejus 175, 177 Philopotamus 2511. Macronema carolina 145, 146 P. corsicanus 25 M. transversum 145, 146 P. ludificatus 25, 27, 89, 91, 165 M. zebratum 145 P. montanus 25, 176 Magellomyia porteri 182 P. variegatus 25 Magnocaricion 111 Philopotamidae 2511., 72 Manophylax 8 Phryganea albifrons 34 Mediterranean islands 7111. P. grandis 89, 90, 175 Membracidae 64, 68 P. interrupta 34 mercury content 161 P. nervosa 35 Mesophylax 71, 89, 105 Phyllolestes (Odon.) 47 M. aspersus 22, 105, 106 phylogeny 5, 711., 5911. M. impunctatus 23, 105, 107 Pileocephalus glyphotaelii (Gregar.) 112 Micrasema 72 Plecoptera 177 Microptema 24, 71, 72, 75, 89, 105 Plectrocnemia 73, 89 M. fissa 105, 112 P. kydon 75 M. lateralis 24, 105, 106 P. renetta 73, 75 M. nycterobia 105, 108, 112, 113 pollution 146, 15911. M. sequax 24, 105, 106, 109 Polycentropus 73 M. testacea 105, 20311. P. flavomaculatus 11811. Molanna angustata 205 P. milikuri 73 M. cinerea 175 Polychaeta 161 Molannidae 175 Potamogeton natans 111 Monocentra improvisa 93 Potamon potamios (Crust.) 75 M. lepidoptera 90, 91 Potamophylax 23, 89, 183, 184 Moselyana 8 P. cingulatus 22, 201 M. comosa 16 P. latipennis 22, 159, 182, 18711. Mystacide 34 P. nigricomis 201 Mystacides 34 P. rotundipennis 22 M. azurea 11811. Potamopyrgus jenkinsi (Moll.) 160 Psychomyidae 29, 72 Neophylax 18 Ptilocolepus 73 Neothremma 42 Pycnopsyche 184 neotropical Trichoptera 47 Nepal 4111. rainbow trout 142, 159 New Caledonia 1 Ranunculus trichophyllus 111 Notidobia 73 Rhyacophila 511., 72, 73, 90, 159 R. angelieri 74 Odonata 177 R. aphrodite 73 Odontocerum 73 R. coloradensis 5 212 R. dorsalis 6, 56 S. mucronatus 105 R. fasciata 5, 163, 164 S. permistus 22, 24, 105 R. glareosa 90, 91 S. sequax 22 R. gudrunae 74 S. vibex speluncarum 105 R. intermedia 6, 89, 91 S.v. vibex 22, 24, 105 R. kelnerae 90, 91 stream zonation 72ft. R. nubila 5, 163, 164 sympatric speciation 62ft. R. pallida 74 Synagapetus 41 R. rougemonti 74 R. tarda 74 Tasimiidae 47 R. torrentium 91 Tasmania 45 R. trifasciata 74 Tassili n' Ajjer 107 R. tristis 91 temperature, influence 106ft., 132ft. R. vulgaris 91 temporary ecosystem 111ft. Rhyacophilidae 29, 47, 90 Thremma 42, 72, 73 Rhyncorheithridae 47 Tinodes 31, 71, 73, 76, 130 Rock River 16711. T. aligi 76 Rytisma acerinum 198 T. kadiellus 75 T. negevianus 75 Salmo gairdneri (Pisces) 159, 161 T. pusillus 142 S. trutta 161 T. raina 76 Sericostoma 19, 89 T. reisseri 76 S. personatum 186 T. waeneri 117ft., 131ft., 153, 154 Sericostomatidae 21ft., 47, 72 T. zelleri 76, 137, 141 Setodes 34 Triaenodes bicolor 149 Silo 18, 73, 89 S. pallipes 163 Uenoa 42 S. nigricomis 149, 163 Uenoidae 41, 42 Simuliidae (Dipt.) 43 Urtica 183 Simulium equinum 159 S. omatum 159 Sisyridae (Neuropt.) 167 vitellogenesis 112 Sjaelland 164 Sortosa 48 Western Alps 87ft. speciation 179 Wormaldia 25ft. spectrography 161 W. copiosa 30, 90, 91, 165 sponge feeding 167 W. mediana 25 Stactobia 76 W. occipitalis 26ft., 91, 164 S. caspersi 76 W. pulla 30 S. mounioti 76 W. subnigra 25 Stenophylacini 105ft. W. triangulifera 26, 31 Stenophylax 23, 24, 71, 76, 89, 105ft., 184 S. crossotus 105 Xenochironomus xenolabris (Dipt.) 167 S. lateralis 22 S. mitis 105 Zygoptera 177

213